% This file was created with JabRef 2.4.2.
% Encoding: UTF8

@ARTICLE{Bailey1982,
  author = {Bailey, T. A. and Dubes, R.},
  title = {Cluster validity profiles},
  journal = {Pattern Recognition},
  year = {1982},
  volume = {15},
  pages = {61-83},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{VanMechelen2004,
  author = {Van Mechelen, I. and Bock, H. H. and De Boeck, P.},
  title = {Two-mode clustering methods: a structured overview},
  journal = {Statistical Methods in Medical Research},
  year = {2004},
  volume = {13},
  pages = {363-394},
  number = {5},
  abstract = {In this paper we present a structured overview of methods for two-mode
	clustering, that is, methods that provide a simultaneous clustering
	of the rows and columns of a rectangular data matrix. Key structuring
	principles include the nature of row, column and data clusters and
	the type of model structure or associated loss function. We illustrate
	with analyses of symptom data on archetypal psychiatric patients.},
  keywords = {ERROR-VARIANCE APPROACH; HIERARCHICAL CLASSES; PROXIMITY DATA; NODAL
	ANALYSIS; ALGORITHM; MODELS; BLOCKMODELS; STRATEGIES},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Tonidandel2004,
  author = {Tonidandel, S. and Overall, J. E.},
  title = {Determining the number of clusters by sampling with replacement},
  journal = {Psychological Methods},
  year = {2004},
  volume = {9},
  pages = {238-249},
  number = {2},
  abstract = {A split-sample replication criterion originally proposed by J. E.
	Overall and K. N. Magee (1992) as a stopping rule for hierarchical
	cluster analysis is applied to multiple data sets generated by sampling
	with replacement from an original simulated primary data set. An
	investigation of the validity of this bootstrap procedure was undertaken
	using different combinations of the true number of latent populations,
	degrees of overlap, and sample sizes. The bootstrap procedure enhanced
	the accuracy of identifying the true number of latent populations
	under virtually all conditions. Increasing the size of the resampled
	data sets relative to the size of the primary data set further increased
	accuracy. A computer program to implement the bootstrap stopping
	rule is made available via a referenced Web site.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Todem2007,
  author = {Todem, D. and Kim, K.},
  title = {Existence and uniqueness conditions for the maximum likelihood solution
	in regression models for correlated Bernoulli data.},
  journal = {Journal of Mathematics and Statistics},
  year = {2007},
  volume = {3},
  pages = {134-141},
  number = {3},
  abstract = {We give sufficient and necessary conditions for the existence of the
	maximum liklihood estimate in a class of multivariate regression
	models for correlated Bernoulli random variables. The models use
	the concept of threshold crossing technique of an underlying multivariate
	latent variable with univariate components formulated as a linear
	regression model. However, in place of their Gaussian assumptions,
	any specified distribution with a strictly increasing cumulative
	distribution function is allowed for error terms. A well known member
	of this class of models is the multivariate probit model. We show
	that our results are a generalization of the concepts of separation
	and overlap of Albert and Anderson for the study of the existence
	of maximum likelihood estimate in generalized linear models. Implications
	of our findings are illustrated through some hypothetical examples.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@INPROCEEDINGS{Tikka2007,
  author = {Tikka, J. and Hollmen, J. and Myllykangas, S.},
  title = {Mixture modeling of DNA copy number amplification patterns in cancer},
  booktitle = {9th International Work-Conference on Artificial Neural Networks},
  year = {2007},
  editor = {Sandoval, F. Prieto A. Cabestany J. Grana M.},
  pages = {972-979},
  address = {San Sebastian, SPAIN},
  abstract = {DNA copy number amplifications are hallmarks of many cancers. In this
	work we analyzed data of genome-wide DNA copy number amplifications
	collected from more than 4500 neoplasm cases. Based on the 0-1 representation
	of the data, we trained finite mixtures of multivariate Bernoulli
	distributions using the EM algorithm to describe the inherent structure
	in the data. The resulting component distributions of the mixtures
	of Bernoulli distributions yielded plausible and localized amplification
	patterns. Individual amplification patterns were tested for their
	role in cancer groups formed with known risk associations. Our detailed
	analysis of chromosome I showed that asbestos-exposure related and
	hormonal imbalance-associated cancers were clustered and specific
	chromosome bands, 1p34 and 1q42, were identified. These sites contain
	cancer genes, which might explain the condition-specific selection
	of these loci for amplification.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Tibshirani2001,
  author = {Tibshirani, R. and Walther, G. and Hastie, T.},
  title = {Estimating the number of clusters in a data set via the gap statistic},
  journal = {Journal of the Royal Statistical Society Series B-Statistical Methodology},
  year = {2001},
  volume = {63},
  pages = {411-423},
  abstract = {We propose a method (the 'gap statistic') for estimating the number
	of clusters (groups) in a set of data. The technique uses the output
	of any clustering algorithm (e.g. K-means or hierarchical), comparing
	the change in within-cluster dispersion with that expected under
	an appropriate reference null distribution. Some theory is developed
	for the proposal and a simulation study shows that the gap statistic
	usually outperforms other methods that have been proposed in the
	literature.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Symons1981,
  author = {Symons, M. J.},
  title = {Clustering criteria and multivariate normal mixtures},
  journal = {Biometrics},
  year = {1981},
  volume = {37},
  pages = {35-43},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Su2001,
  author = {Su, M. S. and Chou, C. H.},
  title = {A modified version of the K-means algorithm with a distance based
	on cluster symmetry},
  journal = {Ieee Transactions on Pattern Analysis and Machine Intelligence},
  year = {2001},
  volume = {23},
  pages = {674-680},
  number = {6},
  abstract = {In this paper, we propose a modified version of the K-means algorithm
	to cluster data. The proposed algorithm adopts a novel nonmetric
	distance measure based on the idea of "point symmetry." This kind
	of "point symmetry distance" can be applied in data clustering and
	human face detection. Several data sets are used to illustrate its
	effectiveness.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Brusco2007a,
  author = {Brusco, M. J. and Steinley, D.},
  title = {A comparison of heuristic procedures for minimum within-cluster sums
	of squares partitioning},
  journal = {Psychometrika},
  year = {2007},
  volume = {72},
  pages = {583-600},
  number = {4},
  abstract = {Perhaps the most common criterion for partitioning a data set is the
	minimization of the within-cluster sums of squared deviation from
	cluster centroids. Although optimal solution procedures for within-cluster
	sums of squares (WCSS) partitioning are computationally feasible
	for small data sets, heuristic procedures are required for most practical
	applications in the behavioral sciences. We compared the performances
	of nine prominent heuristic procedures for WCSS partitioning across
	324 simulated data sets representative of a broad spectrum of test
	conditions. Performance comparisons focused on both percentage deviation
	from the "best-found" WCSS values, as well as recovery of true cluster
	structure. A real-coded genetic algorithm and variable neighborhood
	search heuristic were the most effective methods; however, a straightforward
	two-stage heuristic algorithm, HK-means, also yielded exceptional
	performance. A follow-up experiment using 13 empirical data sets
	from the clustering literature generally supported the results of
	the experiment using simulated data. Our findings have important
	implications for behavioral science researchers, whose theoretical
	conclusions could be adversely affected by poor algorithmic performances.},
  keywords = {combinatorial data analysis; cluster analysis; heuristics; sum of
	squares criterion K-MEANS ALGORITHM; GENETIC ALGORITHM; BINARY DATA;
	VALIDATION; VARIABLES; SELECTION; CRITERIA; BRANCH; SETS},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Strauss1982,
  author = {Strauss, R. E.},
  title = {Statistical significance of species clusters in association analysis},
  journal = {Ecology},
  year = {1982},
  volume = {63},
  pages = {634-639},
  number = {3},
  note = {Times Cited: 48},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Brusco2007,
  author = {Brusco, M. and Steinley, D.},
  title = {A variable neighborhood search method for generalized blockmodeling
	of two-mode binary matrices},
  journal = {Journal of Mathematical Psychology},
  year = {2007},
  volume = {51},
  pages = {325-338},
  number = {5},
  abstract = {The clustering of two-mode proximity matrices is a challenging combinatorial
	optimization problem that has important applications in the quantitative
	social sciences. We focus on one particular type of problem related
	to the clustering of a two-mode binary matrix, which is relevant
	to the establishment of generalized blockmodels for social networks.
	In this context, clusters for the rows of the two-mode matrix intersect
	with clusters of the columns to form blocks, which should ideally
	be either complete (all Is) or null (all Os). A new procedure based
	on variable neighborhood search is presented and compared to an existing
	two-mode K-means clustering algorithm. The new procedure. generally
	provided slightly greater explained variation; however, both methods
	yielded exceptional recovery of cluster structure. (C) 2007 Elsevier
	Inc. All rights reserved.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Steinley2005,
  author = {Steinley, D. and Henson, R.},
  title = {OCLUS: An analytic method for generating clusters with known overlap},
  journal = {Journal of Classification},
  year = {2005},
  volume = {22},
  pages = {221-250},
  number = {2},
  abstract = {The primary method for validating cluster analysis techniques is through
	Monte Carlo simulations that rely on generating data with known cluster
	structure (e.g., Milligan 1996). This paper defines two kinds of
	data generation mechanisms with cluster overlap, marginal and joint;
	current cluster generation methods are framed within these definitions.
	An algorithm generating overlapping clusters based on shared densities
	from several different multivariate distributions is proposed and
	shown to lead to an easily understandable notion of cluster overlap.
	Besides outlining the advantages of generating clusters within this
	framework, a discussion is given of how the proposed data generation
	technique can be used to augment research into current classification
	techniques such as finite mixture modeling, classification algorithm
	robustness, and latent profile analysis.},
  keywords = {cluster generation; overlapping clusters ARTIFICIAL TEST CLUSTERS;
	MIXTURE MODEL TESTS; CLASSIFICATION CAPABILITIES; STOPPING RULES;
	INFORMATION; ALGORITHMS; DISTRIBUTIONS; VARIABLES; SIZE},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Bezdek1998,
  author = {Bezdek, J. C. and Pal, N. R.},
  title = {Some new indexes of cluster validity},
  journal = {Ieee Transactions on Systems Man and Cybernetics Part B-Cybernetics},
  year = {1998},
  volume = {28},
  pages = {301-315},
  number = {3},
  abstract = {We review two clustering algorithms (hard c-means and single linkage)
	and three indexes of crisp cluster validity (Hubert's statistics,
	the Davies-Bouldin index, and Dunn's index). We illustrate two deficiencies
	of Dunn's index which make it overly sensitive to noisy clusters
	and propose several generalizations of it that are not as brittle
	to outliers in the clusters. Our numerical examples show that the
	standard measure of interset distance (the minimum distance between
	points in a pair of sets) is the worst (least reliable) measure upon
	which to base cluster validation indexes when the clusters are expected
	to form volumetric clouds. Experimental results also suggest that
	intercluster separation plays a more important role in cluster validation
	than cluster diameter. Our simulations show that while Dunn's original
	index has operational flaws, the concept it embodies provides a rich
	paradigm for validation of partitions that have cloud-like clusters.
	Five of our generalized Dunn's indexes provide the best validation
	results for the simulations presented.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Steinley2008,
  author = {Steinley, D. and Brusco, M. J.},
  title = {A new variable weighting and selection procedure for {K}-means cluster
	analysis},
  journal = {Multivariate Behavioral Research},
  year = {2008},
  volume = {43},
  pages = {77-108},
  number = {1},
  abstract = {A variance-to-range ratio variable weighting procedure is proposed.
	We show how this weighting method is theoretically grounded in the
	inherent variability found in data exhibiting cluster structure.
	In addition, a variable selection procedure is proposed to operate
	in conjunction with the variable weighting technique. The performances
	of these procedures are demonstrated in a simulation study, showing
	favorable results when compared with existing standardization methods.
	A detailed demonstration of the weighting and selection procedure
	is provided for the well-known Fisher Iris data and several synthetic
	data sets.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Bertrand2006,
  author = {Bertrand, P. and Mufti, G. B.},
  title = {Loevinger's measures of rule quality for assessing cluster stability},
  journal = {Computational Statistics \& Data Analysis},
  year = {2006},
  volume = {50},
  pages = {992-1015},
  number = {4},
  abstract = {A method is developed for measuring clustering stability under the
	removal of a few objects from a set of objects to be partitioned.
	Measures of stability of an individual cluster are defined as Loevinger's
	measures of rule quality. The stability of an individual cluster
	can be interpreted as a weighted mean of the inherent stabilities
	in the isolation and cohesion, respectively, of the examined cluster.
	The design of the method also enables us to measure the stability
	of a partition, that can be viewed as a weighted mean of the stability
	measures of all clusters in the partition. As a consequence, an approach
	is derived for determining the optimal number of clusters of a partition.
	Furthermore, using a Monte Carlo test, a significance probability
	is computed in order to assess how likely any stability measure is,
	under a null model that specifies the absence of cluster stability.
	In order to illustrate the potential of the method, stability measures
	that were obtained by using the batch K-Means algorithm on artificial
	data sets and on Iris Data are presented. (c) 2004 Elsevier B.V.
	All rights reserved.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Bayne1980,
  author = {Bayne, C. K. and Beauchamp, J. J. and Begovich, C. L. and Kane, V.
	E.},
  title = {Monte-{C}arlo comparisons of selected clustering procedures},
  journal = {Pattern Recognition},
  year = {1980},
  volume = {12},
  pages = {51-62},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Steinley2008a,
  author = {Steinley, D. and Brusco, M. J.},
  title = {Selection of variables in cluster analysis: {A}n empirical comparison
	of eight procedures},
  journal = {Psychometrika},
  year = {2008},
  volume = {73},
  pages = {125-144},
  number = {1},
  abstract = {Eight different variable selection techniques for model-based and
	non-model-based clustering are evaluated across a wide range of cluster
	structures. It is shown that several methods have difficulties when
	non-informative variables (i.e., random noise) are included in the
	model. Furthermore, the distribution of the random noise greatly
	impacts the performance of nearly all of the variable selection procedures.
	Overall, a variable selection technique based on a variance-to-range
	weighting procedure coupled with the largest decreases in within-cluster
	sums of squares error performed the best. On the other hand, variable
	selection methods used in conjunction with finite mixture models
	performed the worst.},
  keywords = {cluster analysis; variable selection PROJECTION PURSUIT; LOCAL OPTIMA;
	ALGORITHM},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Steinley2007a,
  author = {Steinley, D. and Brusco, M. J.},
  title = {Initializing {$K$}-means batch clustering: A critical evaluation
	of several techniques},
  journal = {Journal of Classification},
  year = {2007},
  volume = {24},
  pages = {99-121},
  number = {1},
  abstract = {K-means clustering is arguably the most popular technique for partitioning
	data. Unfortunately, K-means suffers from the well-known problem
	of locally optimal solutions. Furthermore, the final partition is
	dependent upon the initial configuration, making the choice of starting
	partitions all the more important. This paper evaluates 12 procedures
	proposed in the literature and provides recommendations for best
	practices.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Steinley2007,
  author = {Steinley, D.},
  title = {Validating clusters with the lower bound for sum-of-squares error},
  journal = {Psychometrika},
  year = {2007},
  volume = {72},
  pages = {93-106},
  number = {1},
  abstract = {Given that a minor condition holds (e.g., the number of variables
	is greater than the number of clusters), a nontrivial lower bound
	for the sum-of-squares error criterion in K-means clustering is derived.
	By calculating the lower bound for several different situations,
	a method is developed to determine the adequacy of cluster solution
	based on the observed sum-of-squares error as compared to the minimum
	sum-of-squares error.},
  keywords = {k-means; cluster analysis ARTIFICIAL TEST CLUSTERS; DATA SET; CLASSIFICATION;
	ALGORITHM; NUMBER},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Steinley2006,
  author = {Steinley, D.},
  title = {Profiling local optima in {$K$}-means clustering: {D}eveloping a
	diagnostic techniques},
  journal = {Psychological Methods},
  year = {2006},
  volume = {11},
  pages = {178-192},
  number = {2},
  abstract = {Using the cluster generation procedure proposed by D. Steinley and
	R. Henson (2005), the author investigated the performance of K-means
	clustering under the following scenarios: (a) different probabilities
	of cluster overlap; (b) different types of cluster overlap; (c) varying
	samples sizes, clusters, and dimensions; (d) different multivariate
	distributions of clusters; and (e) various multidimensional data
	structures. The results are evaluated in terms of the Hubert-Arabie
	adjusted Rand index, and several observations concerning the performance
	of K-means clustering are made. Finally, the article concludes with
	the proposal of a diagnostic technique indicating when the partitioning
	given by a K-means cluster analysis can be trusted. By combining
	the information from several observable characteristics of the data
	(number of clusters, number of variables, sample size, etc.) with
	the prevalence of unique local optima in several thousand implementations
	of the K-means algorithm, the author provides a method capable of
	guiding key data-analysis decisions.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Steinley2006a,
  author = {Steinley, D.},
  title = {{$K$}-means clustering: {A} half-century synthesis},
  journal = {British Journal of Mathematical \& Statistical Psychology},
  year = {2006},
  volume = {59},
  pages = {1-34},
  abstract = {This paper synthesizes the results, methodology, and research conducted
	concerning the K-means clustering method over the last fifty years.
	The K-means method is first introduced, various formulations of the
	minimum variance loss function and alternative loss functions within
	the same class are outlined, and different methods of choosing the
	number of clusters and initialization, variable preprocessing, and
	data reduction schemes are discussed. Theoretic statistical results
	are provided and various extensions of K-means using different metrics
	or modifications of the original algorithm are given, leading to
	a unifying treatment of K-means and some of its extensions. Finally,
	several future studies are outlined that could enhance the understanding
	of numerous subtleties affecting the performance of the K-means method.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Steinley2004,
  author = {Steinley, D.},
  title = {Properties of the {H}ubert-{A}rabie adjusted {R}and index},
  journal = {Psychological Methods},
  year = {2004},
  volume = {9},
  pages = {386-396},
  number = {3},
  abstract = {This article provides an investigation of cluster validation indices
	that relates 4 of the indices to the L. Hubert and P. Arabie (1985)
	adjusted Rand index-the cluster validation measure of choice (G.
	W. Milligan & M. C. Cooper, 1986). It is shown how these other indices
	can be "roughly" transformed into the same scale as the adjusted
	Rand index. Furthermore, in-depth explanations are given of why classification
	rates should not be used in cluster validation research. The article
	concludes by summarizing several properties of the adjusted Rand
	index across many conditions and provides a method for testing the
	significance of observed adjusted Rand indices.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Steinley2004a,
  author = {Steinley, D.},
  title = {Standardizing variables in {$K$}-means clustering},
  journal = {Classification, Clustering, and Data Mining Applications},
  year = {2004},
  pages = {53-60},
  note = {Banks, D House, L McMorris, FR Arabie, P Gaul, W Meeting of the International-Federation-of-Classifications-Societies
	(IFCS) JUL 15-18, 2004 Illinois Inst Technol, Chicago, IL},
  abstract = {Several standardization methods are investigated in conjunction with
	the K-means algorithm under various conditions. We find that traditional
	standardization methods (i.e., z-scores) are inferior to alternative
	standardization methods. Future suggestions concerning the combination
	of standardization and variable selection are considered.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Steinley2003,
  author = {Steinley, D.},
  title = {Local optima in {$K$}-means clustering: {W}hat you don't know may
	hurt you},
  journal = {Psychological Methods},
  year = {2003},
  volume = {8},
  pages = {294-304},
  number = {3},
  abstract = {The popular K-means clustering method, as implemented in 3 commercial
	software packages (SPSS, SYSTAT, and SAS), generally provides solutions
	that are only locally optimal for a given set of data. Because none
	of these commercial implementations offer a reasonable mechanism
	to begin the K-means method at alternative starting points, separate
	routines were written within the MATLAB (MathWorks, 1999) environment
	that can be initialized randomly (these routines are provided at
	the end of the online version of this article in the PsycARTICLES
	database). Through the analysis of 2 empirical data sets and 8 10
	simulated data sets, it is shown that the results provided by commercial
	packages are most likely locally optimal. These results suggest the
	need for some strategy to study the local optima problem for a specific
	data set or to identify methods for finding "good" starting values
	that might lead to the best solutions possible.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@INPROCEEDINGS{Song2006,
  author = {Song, S. and Li, C. P.},
  title = {Improved ROCK for text clustering using asymmetric proximity},
  booktitle = {32nd Conference on Current Trends in Theory and Practice of Computer
	Science},
  year = {2006},
  editor = {Wiedermann, J. Tel G. Pokorny J. Bielikova M. Stuller J.},
  pages = {501-510},
  address = {Merin, CZECH REPUBLIC},
  abstract = {The ROCK algorithm can be applied to text clustering in large databases.
	The effectiveness of ROCK, however, is limited, because of the high
	dimensionality of textual data and traditional proximity measure
	of documents. In this paper, we propose an improved approach to strengthen
	the discriminative feature of text documents, which uses asymmetric
	proximity. Instead of the links count in ROCK, we propose a novel
	concept of link weight overlaps to measure the proximity between
	two clusters. The IROCK (Improved ROCK) algorithm performs clustering
	analysis based on the overlap information of asymmetric proximities
	between text objects. We carry on the clustering process in an agglomerative
	hierarchical way. To demonstrate the effectiveness of IROCK, we perform
	an experimental evaluation on real textual data. A comparison with
	ROCK and classical algorithms indicates the superiority of our approach.},
  keywords = {data mining; text clustering},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Snijders1991,
  author = {Snijders, T. A. B.},
  title = {Enumeration and simulation methods for 0-1 matrices with given marginals},
  journal = {Psychometrika},
  year = {1991},
  volume = {56},
  pages = {397-417},
  number = {3},
  note = {Times Cited: 38},
  abstract = {Data in the form of zero-one matrices where conditioning on the marginals
	is relevant arise in diverse fields such as social networks and ecology;
	directed graphs constitute an important special case. An algorithm
	is given for the complete enumeration of the family of all zero-one
	matrices with given marginals and with a prespecified set of cells
	with structural zero entries. Complete enumeration is computationally
	feasible only for relatively small matrices. Therefore, a more useable
	Monte Carlo simulation method for the uniform distribution over this
	family is given, based on unequal probability sampling and ratio
	estimation. This method is applied to testing reciprocity of choices
	in social networks.},
  keywords = {ADJACENCY MATRICES; RANDOM DIGRAPHS, NETWORKS; ECOLOGY; MONTE-CARLO
	METHODS; UNEQUAL PROBABILITY SAMPLING; RECIPROCITY DISTRIBUTIONS;
	NETWORKS; CENSUS},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Shatovska2008,
  author = {Shatovska},
  title = {The new software package for dynamic hierarchical clustering for
	circles types of shapes},
  year = {2008},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@INPROCEEDINGS{Seppanen2003,
  author = {Seppanen, J. K. and Bingham, E. and Mannila, H.},
  title = {A simple algorithm for topic identification in 0-1 data},
  booktitle = {7th European Conference on Principles and Practice of Knowledge Discovery
	in Databases},
  year = {2003},
  editor = {Lavrac, N. Gamberger D. Todorovski L. Blockeel H.},
  pages = {423-434},
  address = {Cavtat, Croatia},
  abstract = {Topics in 0-1 datasets are sets of variables whose occurrences are
	positively connected together. Earlier, we described a simple generative
	topic model. In this paper we show that, given data produced by this
	model, the lift statistics of attributes can be described in matrix
	form. We use this result to obtain a simple algorithm for finding
	topics in 0-1 data. We also show that a problem related to the identification
	of topics is NP-hard. We give experimental results on the topic identification
	problem, both on generated and real data.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Scott1999,
  author = {Scott, P. D. and Wilkins, E.},
  title = {Evaluating data mining procedures: techniques for generating artificial
	data sets},
  journal = {Information and Software Technology},
  year = {1999},
  volume = {41},
  pages = {579-587},
  number = {9},
  abstract = {In this article, we discuss the need to evaluate the performance of
	data mining procedures and argue that tests done with real data sets
	cannot provide all the information needed for a thorough assessment
	of their performance characteristics. We argue that artificial data
	sets are therefore essential. After a discussion of the desirable
	characteristics of such artificial data, we describe two pseudo-random
	generators. The first is based on the multi-variate normal distribution
	and gives the investigator full control of the degree of correlation
	between the variables in the artificial data sets. The second is
	inspired by fractal techniques for synthesizing artificial landscapes
	and can produce data whose classification complexity can be controlled
	by a single parameter. We conclude with a discussion of the additional
	work necessary to achieve the ultimate goal of a method of matching
	data sets to the most appropriate data mining technique. (C) 1999
	Elsevier Science B.V. All rights reserved.},
  keywords = {data mining; artificial data sets; pseudo-random generators},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Scott1971,
  author = {Scott, A. J. and Symons, M. J.},
  title = {Clustering methods based on likelihood ratio criteria},
  journal = {Biometrics},
  year = {1971},
  volume = {27},
  pages = {387-\&},
  number = {2},
  note = {Times Cited: 111},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Savicky2006,
  author = {Savicky},
  title = {Problems with Matlab RNG},
  year = {2006},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Real1996,
  author = {Real, R. and Vargas, J. M.},
  title = {The probabilistic basis of Jaccard's index of similarity},
  journal = {Systematic Biology},
  year = {1996},
  volume = {45},
  pages = {380-385},
  number = {3},
  note = {Times Cited: 15},
  keywords = {SPECIES CO-OCCURRENCES; NULL MODELS; INTERSPECIFIC COMPETITION; OCCURRENCE
	PATTERNS; BINARY DATA; BIOGEOGRAPHY; ISLANDS; COMMUNITIES},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Qiu2006,
  author = {Qiu, W. L. and Joe, H.},
  title = {Generation of random clusters with specified degree of separation},
  journal = {Journal of Classification},
  year = {2006},
  volume = {23},
  pages = {315-334},
  number = {2},
  abstract = {We propose a random cluster generation algorithm that has the desired
	features: (1) the population degree of separation between clusters
	and the nearest neighboring clusters can be set to a specified value,
	based on a separation index; (2) no constraint is imposed on the
	isolation among clusters in each dimension; (3) the covariance matrices
	correspond to different shapes, diameters and orientations; (4) the
	full cluster structures generally could not be detected simply from
	pair-wise scatterplots of variables; (5) noisy variables and outliers
	can be imposed to make the cluster structures harder to be recovered.
	This algorithm is an improvement on the method used in Milligan (1985).},
  keywords = {cluster generation; separation index; factorial experiment design
	ARTIFICIAL TEST CLUSTERS; ALGORITHM; VARIABLES; SELECTION; INDEX},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Potvin1993,
  author = {Potvin, C. and Roff, D. A.},
  title = {Distribution-free and robust statistical-methods - viable alternatives
	to parametric statistics},
  journal = {Ecology},
  year = {1993},
  volume = {74},
  pages = {1617-1628},
  number = {6},
  note = {Times Cited: 255},
  abstract = {After making a case for the prevalence of nonnormality, this paper
	attempts to introduce some distribution-free and robust techniques
	to ecologists and to offer a critical appraisal of the potential
	advantages and drawbacks of these methods. The techniques presented
	fall into two distinct categories, methods based on ranks and ''computer-intensive''
	techniques. Distribution-free rank tests have features that can be
	recommended. They free the practitioner from concern about the underlying
	distribution and are very robust to outliers. If the distribution
	underlying the observations is other than normal, rank tests tend
	to be more efficient than their parametric counterparts. The absence,
	in computing packages, of rank procedures for complex designs may,
	however, severely limit their use for ecological data. An entire
	body of novel distribution-free methods has been developed in parallel
	with the increasing capacities of today's computers to process large
	quantities of data. These techniques either reshuffle or resample
	a data set (i.e., sample with replacement) in order to perform their
	analyses. The former we shall refer to as ''permutation'' or ''randomization''
	methods and the latter as ''bootstrap'' techniques. These computer-intensive
	methods provide new alternatives for the problem of a small and/or
	unbalanced data set, and they may be the solution for parameter estimation
	when the sampling distribution cannot be derived analytically. Caution
	must be exercised in the interpretation of these estimates because
	confidence limits may be too small.},
  keywords = {POPULATION-GROWTH RATES; NONPARAMETRIC STATISTICS; MITOCHONDRIAL-DNA;
	ANIMAL NUMBERS; WILLIAMS-TEST; MONTE-CARLO; MANTEL TEST; DOSE LEVELS;
	REGRESSION; DESIGNS},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Park1996,
  author = {Park, C. G. and Park, T. and Shin, D. W.},
  title = {A simple method for generating correlated binary variates},
  journal = {American Statistician},
  year = {1996},
  volume = {50},
  pages = {306-310},
  number = {4},
  abstract = {Correlated binary data are frequently analyzed in studies of repeated
	measurements, reliability analysis, and others. In such studies correlations
	among binary variables are usually nonnegative. This article provides
	a simple algorithm for generating an arbitrary dimensional random
	vector of nonnegatively correlated binary variables, Ln some frequently
	encountered situations the algorithm reduces to explicit expressions.
	The correlated binary variables are generated from correlated Poisson
	variables, The key idea lies in the property that any Poisson random
	variable can be expressed as a convolution of other independent Poisson
	random variables, The binary variables have desired correlations
	by sharing common independent Poisson variables.},
  keywords = {generalized estimating equations; Poisson variables; random number
	generation LONGITUDINAL DATA; MODELS},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Panayirci1983,
  author = {Panayirci, E. and Dubes, R. C.},
  title = {A test for multidimensional clustering tendency},
  journal = {Pattern Recognition},
  year = {1983},
  volume = {16},
  pages = {433-444},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Joe2006,
  author = {Joe, H.},
  title = {Generating random correlation matrices based on partial correlations},
  journal = {Journal of Multivariate Analysis},
  year = {2006},
  volume = {97},
  pages = {2177-2189},
  number = {10},
  abstract = {A d-dimensional positive definite correlation matrix R = (rho(ij))
	can be parametrized in terms of the correlations rho(i,i+1) for i
	= 1,..., d - 1, and the partial correlations rho(ij\i+1,.... j-1)
	for j - i >= 2. These ((d)(2)) parameters can independently take
	values in the interval (- 1, 1). Hence we can generate a random positive
	definite correlation matrix by choosing independent distributions
	F-ij, 1 <= i < j <= d, for these ((d)(2)) parameters. We obtain conditions
	on the F-ij so that the joint density of (rho(ij)) is proportional
	to a power of det(R) and hence independent of the order of indices
	defining the sequence of partial correlations. As a special case,
	we have a simple construction for generating R that is uniform over
	the space of positive definite correlation matrices. As a byproduct,
	we determine the volume of the set of correlation matrices in ((d)(2))-dimensional
	space. To prove our results, we obtain a simple remarkable identity
	which expresses det(R) as a function of rho(i,i+1) for i = 1,...,
	d - 1, and p(ij\i+1,... j-1) for j - i >= 2. (C) 2005 Elsevier Inc.
	All rights reserved.},
  keywords = {beta distribution; determinant of correlation matrix},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@INPROCEEDINGS{Joe2004,
  author = {Joe, H.},
  title = {Range of correlation matrices for dependent random variables with
	given marginal distributions},
  booktitle = {International Conference on Distribution Theory, Order Statistics,
	and Inference},
  year = {2004},
  editor = {Balakrishnan, N. Castillo E. Sarabia J. M.},
  pages = {125-142},
  address = {Santander, SPAIN},
  abstract = {Let X-1, center dot center dot center dot, X-d be d (d >= 3) dependent
	random variables with finite variances such that X-j similar to F-j.
	Results on the set S-d(F-1, center dot center dot center dot, F-d)
	of possible correlation matrices with given margins are obtained;
	this set is relevant for simulating dependent random variables with
	given marginal distributions and a given correlation matrix. When
	F-1 = (...) = F-d = F, we let S-d(F) denote the set of possible correlation
	matrices. Of interest is the set of F for which Sd(F) is the same
	as the set of all non-negative definite correlation matrices; using
	a construction with conditional distributions, we show that this
	property holds only if F is a (location-scale shift of a) margin
	of a (d-1)-dimensional spherical distribution.},
  keywords = {spherically symmetric; elliptically contoured; copula; partial correlation;
	Frechet bounds},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Jaksic1990,
  author = {Jaksic, F. M. and Medel, R. G.},
  title = {Objective recognition of guilds - testing for statistically significant
	species clusters},
  journal = {Oecologia},
  year = {1990},
  volume = {82},
  pages = {87-92},
  number = {1},
  note = {Times Cited: 52},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Jain1987,
  author = {Jain, A. K. and Moreau, J. V.},
  title = {Bootstrap technique in cluster-analysis},
  journal = {Pattern Recognition},
  year = {1987},
  volume = {20},
  pages = {547-568},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Jackson1992,
  author = {Jackson, D. A. and Somers, K. M. and Harvey, H. H.},
  title = {Null models and fish communities - evidence of nonrandom patterns},
  journal = {American Naturalist},
  year = {1992},
  volume = {139},
  pages = {930-951},
  number = {5},
  note = {Times Cited: 62},
  abstract = {Studies of ecological communities often make implicit assumptions
	that the species have nonrandom patterns organized through biotic
	and abiotic factors. Although such assumptions are generally not
	tested, the analyses and conclusions derived depend on nonrandom
	patterns' being present. Several "null" or "neutral" models have
	been proposed to test for these patterns. We contrast two of the
	more prevalent models and develop two new models, subsequently evaluating
	them with five sets of fish community data. Three of the null models
	provide similar results, from which it is concluded that fish communities
	from five regions of Ontario are nonrandomly structured. These three
	models evaluate pairs of species according to departures from null
	or random co-occurrence expectations, A Monte Carlo model based on
	the procedure proposed by E. F. Connor and D. Simberloff supports
	random community organization, but we attribute this discrepancy
	to the conservative pooling of species-pair information in that model.
	We recommend a hybrid model combining Monte Carlo and log-linear
	methods for future studies, although the log-linear model of M. E.
	Gilpin and J. M. Diamond provides a reasonable approximation. On
	the basis of species associations derived from the various models,
	we attribute much of the nonrandom structure to common habitat requirements
	among co-occurring species. A predominance of positively associated
	species generally involves pairs of species with similar ecological
	characteristics. Strong negative associations typically involve predator-prey
	species. Although competition is often identified as a significant
	feature in community ecology, we do not believe that competition
	is a major force structuring these fish communities.},
  keywords = {SPECIES CO-OCCURRENCES; INTERSPECIFIC COMPETITION; LAKES; ASSEMBLAGES;
	SIMILARITY; ISLANDS; STOCHASTICITY; ASSOCIATION; DIVERSITY; ECOLOGY},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Jackson1989,
  author = {Jackson, D. A. and Somers, K. M. and Harvey, H. H.},
  title = {Similarity coefficients - measures of co-occurrence and association
	or simply measures of occurrence},
  journal = {American Naturalist},
  year = {1989},
  volume = {133},
  pages = {436-453},
  number = {3},
  note = {Times Cited: 88},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Irigoien2008,
  author = {Irigoien, I. and Arenas, C.},
  title = {INCA: New statistic for estimating the number of clusters and identifying
	atypical units},
  journal = {Statistics in Medicine},
  year = {2008},
  volume = {27},
  pages = {2948-2973},
  number = {15},
  abstract = {This paper presents a solution to two problems that arise in the classification
	of data such as types of tumor, samples of gene expression profiles
	or general biomedical data. First, to estimate the real number of
	clusters in a data set and second to decide whether a new unit belongs
	to one of these previously identified clusters or it is an outlier
	or atypical unit. We propose a new statistic which allows us to solve
	these problems. As our approach is based on a measure of distance
	or dissimilarity between any pair of units, it can be applied to
	any kind of multivariate data (continuous, binary or multi-attribute
	data) and it has applications in many biomedical fields. We validated
	the approach in simulated examples and applied it to the diagnosis
	of dermal diseases and to the analysis of lymphatic cancer data,
	showing the good performance of our approach. Copyright (C) 2007
	John Wiley & Sons, Ltd.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Hunter2004,
  author = {Hunter, J. C. and McCoy, R. A.},
  title = {Applying randomization tests to cluster analyses},
  journal = {Journal of Vegetation Science},
  year = {2004},
  volume = {15},
  pages = {135-138},
  number = {1},
  note = {Times Cited: 0},
  abstract = {In applying randomization tests to hierarchical cluster analyses,
	we have noted a potentially misleading result: within a significant
	group. linkages are often identified as significant even when species
	are randomly distributed among the group's sites. We demonstrate
	this through a cluster analysis of a constructed matrix with two
	groups of 20 sites that share no species, and within each group species
	are randomly distributed among sites. A randomization test identified
	both of the groups and all linkages within them as significant, while
	the same test found all linkages non-significant in the cluster analysis
	of a matrix containing just one of the two groups of 20 sites. In
	general, a non-random distribution of species within a data set shortens
	linkages relative to distances in null distributions derived from
	randomized versions of the data. This confounds efforts to identify
	significant sub-groups within a significant group. However. the significance
	of sub-groups possibly could be tested by comparing linkage distances
	to a null distribution derived from the randomization and clustering
	of a sub-matrix containing only the sites within the larger group.
	In essence, this comparison tests the null hypothesis that within
	the significant group. sites represent random assemblages of species.
	When applied to actual data sets. an approach involving sequential
	randomization tests could allow the evaluation of all nodes in a
	classification. increasing, the utility of randomization tests and
	strengthening the interpretation of groups produced by cluster analysis.},
  keywords = {classification; null distribution; permutation test; resampling; vegetation
	SPECIES CLUSTERS},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Hubert1976a,
  author = {Hubert, L. J. and Levin, J. R.},
  title = {General statistical framework for assessing categorical clustering
	in free-recall},
  journal = {Psychological Bulletin},
  year = {1976},
  volume = {83},
  pages = {1072-1080},
  number = {6},
  note = {Times Cited: 52},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Hubert1976,
  author = {Hubert, L. and Schultz, J.},
  title = {Quadratic assignment as a general data-analysis strategy},
  journal = {British Journal of Mathematical \& Statistical Psychology},
  year = {1976},
  volume = {29},
  pages = {190-241},
  number = {NOV},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Orloci1967,
  author = {Orloci, L.},
  title = {An agglomerative method for classification of plant communities},
  journal = {Journal of Ecology},
  year = {1967},
  volume = {55},
  pages = {193-\&},
  number = {1},
  note = {Times Cited: 244},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Baroni-Urbani1976,
  author = {Baroni-Urbani, Cesare and Buser, Mauro W.},
  title = {Similarity of Binary Data},
  journal = {Systematic Zoology},
  year = {1976},
  volume = {25},
  pages = {251--259},
  number = {3},
  abstract = {A set of intuitively obvious properties of a coefficient of similarity
	for binary data is established. Critical examination of the coefficients
	available from the literature shows that none of them satisfies all
	these properties, and a new coefficient is proposed to obviate this
	inconvenience. The distribution of the new coefficient is studied
	on very large, perfectly random samples of OTU's, and a table is
	constructed to show its critical values at different significance
	levels. A mathematical appendix is given to demonstrate the procedure
	used to construct perfectly random samples of different size.},
  copyright = {Copyright © 1976 Society of Systematic Biologists},
  issn = {00397989},
  jstor_articletype = {primary_article},
  jstor_formatteddate = {Sep., 1976},
  owner = {rebeccaspeckman},
  publisher = {Taylor \& Francis, Ltd. for the Society of Systematic Biologists},
  timestamp = {2009.03.03}
}

@ARTICLE{Nemec1988,
  author = {Nemec, A. F. L. and Brinkhurst, R. O.},
  title = {Using the bootstrap to assess statistical significance in the cluster-analysis
	of species abundance data},
  journal = {Canadian Journal of Fisheries and Aquatic Sciences},
  year = {1988},
  volume = {45},
  pages = {965-970},
  number = {6},
  note = {Times Cited: 31},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Mojena1977,
  author = {Mojena, R.},
  title = {Hierarchical grouping methods and stopping rules: an evaluation},
  journal = {Computer Journal},
  year = {1977},
  volume = {20},
  pages = {359-363},
  number = {4},
  note = {Times Cited: 207},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Milligan1983,
  author = {Milligan, G. W. and Soon, S. C. and Sokol, L. M.},
  title = {The effect of cluster size, dimensionality, and the number of clusters
	on recovery of true cluster structure},
  journal = {Ieee Transactions on Pattern Analysis and Machine Intelligence},
  year = {1983},
  volume = {5},
  pages = {40-47},
  number = {1},
  note = {Times Cited: 57},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Milligan1987,
  author = {Milligan, G. W. and Cooper, M. C.},
  title = {Methodology review - clustering methods},
  journal = {Applied Psychological Measurement},
  year = {1987},
  volume = {11},
  pages = {329-354},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Milligan1985,
  author = {Milligan, G. W. and Cooper, M. C.},
  title = {An examination of procedures for determining the number of clusters
	in a data set},
  journal = {Psychometrika},
  year = {1985},
  volume = {50},
  pages = {159-179},
  number = {2},
  note = {I have hard copy},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Milligan1981,
  author = {Milligan, G. W.},
  title = {A {M}onte-{C}arlo study of 30 internal criterion measures for cluster
	analysis},
  journal = {Psychometrika},
  year = {1981},
  volume = {46},
  pages = {187-199},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Milligan1981a,
  author = {Milligan, G. W.},
  title = {A review of {M}onte-{C}arlo tests of cluster analysis},
  journal = {Multivariate Behavioral Research},
  year = {1981},
  volume = {16},
  pages = {379-407},
  number = {3},
  note = {Times Cited: 114},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{McShane2002,
  author = {McShane, L. M. and Radmacher, M. D. and Freidlin, B. and Yu, R. and
	Li, M. C. and Simon, R.},
  title = {Methods for assessing reproducibility of clustering patterns observed
	in analyses of microarray data},
  journal = {Bioinformatics},
  year = {2002},
  volume = {18},
  pages = {1462-1469},
  number = {11},
  note = {I have hard copy},
  abstract = {Motivation: Recent technological advances such as cDNA microarray
	technology have made it possible to simultaneously interrogate thousands
	of genes in a biological specimen. A cDNA microarray experiment produces
	a gene expression 'profile'. Often interest lies in discovering novel
	subgroupings, or 'clusters', of specimens based on their profiles,
	for example identification of new tumor taxonomies. Cluster analysis
	techniques such as hierarchical clustering and self-organizing maps
	have frequently been used for investigating structure in microarray
	data. However, clustering algorithms always detect clusters, even
	on random data, and it is easy to misinterpret the results without
	some objective measure of the reproducibility of the clusters. Results:
	We present statistical methods for testing for overall clustering
	of gene expression profiles, and we define easily interpretable measures
	of cluster-specific reproducibility that facilitate understanding
	of the clustering structure. We apply these methods to elucidate
	structure in cDNA microarray gene expression profiles obtained on
	melanoma tumors and on prostate specimens.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Huang1998,
  author = {Huang, Z. X.},
  title = {Extensions to the $k$-{M}eans algorithm for clustering large data
	sets with categorical values},
  journal = {Data Mining and Knowledge Discovery},
  year = {1998},
  volume = {2},
  pages = {283-304},
  number = {3},
  abstract = {The k-means algorithm is well known for its efficiency in clustering
	large data sets. However, working only on numeric values prohibits
	it from being used to cluster real world data containing categorical
	values. In this paper we present two algorithms which extend the
	k-means algorithm to categorical domains and domains with mixed numeric
	and categorical values. The k-modes algorithm uses a simple matching
	dissimilarity measure to deal with categorical objects, replaces
	the means of clusters with modes, and uses a frequency-based method
	to update modes in the clustering process to minimise the clustering
	cost function. With these extensions the k-modes algorithm enables
	the clustering of categorical data in a fashion similar to k-means.
	The k-prototypes algorithm, through the definition of a combined
	dissimilarity measure, further integrates the k-means and k-modes
	algorithms to allow for clustering objects described by mixed numeric
	and categorical attributes. We use the well known soybean disease
	and credit approval data sets to demonstrate the clustering performance
	of the two algorithms. Our experiments on two real world data sets
	with half a million objects each show that the two algorithms are
	efficient when clustering large data sets, which is critical to data
	mining applications.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{He2008,
  author = {He, Z. Y. and Xu, X. F. and Deng, S. C.},
  title = {k-ANMI: A mutual information based clustering algorithm for categorical
	data},
  journal = {Information Fusion},
  year = {2008},
  volume = {9},
  pages = {223-233},
  number = {2},
  abstract = {Clustering categorical data is an integral part of data mining and
	has attracted much attention recently. In this paper, we present
	k-ANMI, a new efficient algorithm for clustering categorical data.
	The k-ANMI algorithm works in a way that is similar to the popular
	k-means algorithm, and the goodness of clustering in each step is
	evaluated using a mutual information based criterion (namely, average
	normalized mutual information-ANMI) borrowed from cluster ensemble.
	This algorithm is easy to implement, requiring multiple hash tables
	as the only major data structure. Experimental results on real datasets
	show that k-ANMI algorithm is competitive with those state-of-the-art
	categorical data clustering algorithms with respect to clustering
	accuracy. (c) 2006 Elsevier B.V. All rights reserved.},
  keywords = {clustering; categorical data; mutual information; cluster ensemble;
	data mining DISSIMILARITY MEASURE; SYMBOLIC PATTERNS; DATA SETS},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{He2005,
  author = {He, Z. Y. and Deng, S. C. and Xu, X. F.},
  title = {Improving K-modes algorithm considering frequencies of attribute
	values in mode},
  journal = {Computational Intelligence and Security, Pt 1, Proceedings},
  year = {2005},
  volume = {3801},
  pages = {157-162},
  note = {International Conference on Computational Intelligence and Security
	DEC 15-19, 2005 Xi'an, PEOPLES R CHINA},
  abstract = {In this paper, we present an experimental study on applying a new
	dissimilarity measure to the k-modes clustering algorithm to improve
	its clustering accuracy. The measure is based on the idea that the
	similarity between a data object and cluster mode, is directly proportional
	to the sum of relative frequencies of the common values in mode.
	Experimental results on real life datasets show that, the modified
	algorithm is superior to the original k-modes algorithm with respect
	to clustering accuracy.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Handl2005,
  author = {Handl, J. and Knowles, J. and Kell, D. B.},
  title = {Computational cluster validation in post-genomic data analysis},
  journal = {Bioinformatics},
  year = {2005},
  volume = {21},
  pages = {3201-3212},
  number = {15},
  abstract = {Motivation: The discovery of novel biological knowledge from the ab
	initio analysis of post-genomic data relies upon the use of unsupervised
	processing methods, in particular clustering techniques. Much recent
	research in bioinformatics has therefore been focused on the transfer
	of clustering methods introduced in other scientific fields and on
	the development of novel algorithms specifically designed to tackle
	the challenges posed by post-genomic data. The partitions returned
	by a clustering algorithm are commonly validated using visual inspection
	and concordance with prior biological knowledge-whether the clusters
	actually correspond to the real structure in the data is somewhat
	less frequently considered. Suitable computational cluster validation
	techniques are available in the general data-mining literature, but
	have been given only a fraction of the same attention in bioinformatics.
	Results: This review paper aims to familiarize the reader with the
	battery of techniques available for the validation of clustering
	results, with a particular focus on their application to post-genomic
	data analysis. Synthetic and real biological datasets are used to
	demonstrate the benefits, and also some of the perils, of analytical
	clustervalidation. Availability: The software used in the experiments
	is available at http://dbkweb.ch.umist.ac.uk/handl/clustervalidation/
	Contact: J.Handl@postgrad.manchester.ac.uk Supplementary information:
	Enlarged colour plots are provided in the Supplementary Material,
	which is available at http://dbkweb.ch.umist.ac.uk/handl/clustervalidation/},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Halkidi,
  author = {Halkidi},
  title = {clustering validation review-- no date or journal -- they have stuff
	in journals},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Guvenir1998,
  author = {Guvenir, H. A. and Demiroz, G. and Ilter, N.},
  title = {Learning differential diagnosis of erythemato-squamous diseases using
	voting feature intervals},
  journal = {Artificial Intelligence in Medicine},
  year = {1998},
  volume = {13},
  pages = {147-165},
  number = {3},
  abstract = {A new classification algorithm, called VFI5 (for Voting Feature Intervals),
	is developed and applied to problem of differential diagnosis of
	erythemato-squamous diseases. The domain contains records of patients
	with known diagnosis. Given a training set of such records, the VFI5
	classifier learns how to differentiate a new case in the domain.
	VFI5 represents a concept in the form of feature intervals on each
	feature dimension separately. classification in the VFI5 algorithm
	is based on a real-valued voting. Each feature equally participates
	in the voting process and the class that receives the maximum amount
	of votes is declared to be the predicted class. The performance of
	the VFI5 classifier is evaluated empirically in terms of classification
	accuracy and running time. (C) 1998 Elsevier Science B.V. All rights
	reserved.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Guha2000,
  author = {Guha, S. and Rastogi, R. and Shim, K.},
  title = {Rock: A robust clustering algorithm for categorical attributes},
  journal = {Information Systems},
  year = {2000},
  volume = {25},
  pages = {345-366},
  number = {5},
  abstract = {Clustering, in data mining, is useful to discover distribution patterns
	in the underlying data. Clustering algorithms usually employ a distance
	metric based (e.g., euclidean) similarity measure in order to partition
	the database such that data points in the same partition are more
	similar than points in different partitions. In this paper, we study
	clustering algorithms for data with boolean and categorical attributes.
	We show that traditional clustering algorithms that use distances
	between points for clustering are not appropriate for boolean and
	categorical attributes. Instead, we propose a novel concept of links
	to measure the similarity/proximity between a pair of data points.
	We develop a robust hierarchical clustering algorithm ROCK that employs
	links and not distances when merging clusters. Our methods naturally
	extend to non-metric similarity measures that are relevant in situations
	where a domain expert/similarity table is the only source of knowledge.
	In addition to presenting detailed complexity results for ROCK, we
	also conduct an experimental study with real-life as well as synthetic
	data sets to demonstrate the effectiveness of our techniques. For
	data with categorical attributes, our findings indicate that ROCK
	not only generates better quality clusters than traditional algorithms,
	but it also exhibits good scalability properties. (C) 2000 Published
	by Elsevier Science Ltd.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@INPROCEEDINGS{Grim2007,
  author = {Grim, J. and Hora, J.},
  title = {Minimum information loss cluster analysis for categorical data},
  booktitle = {5th International Conference on Machine Learning and Data Mining
	in Pattern Recognition},
  year = {2007},
  editor = {Perner, P.},
  pages = {233-247},
  address = {Leipzig, GERMANY},
  abstract = {The EM algorithm has been used repeatedly to identify latent classes
	in categorical data by estimating finite distribution mixtures of
	product components. Unfortunately, the underlying mixtures are not
	uniquely identifiable and, moreover, the estimated mixture parameters
	are starting-point dependent. For this reason we use the latent class
	model only to define a set of "elementary" classes by estimating
	a mixture of a large number components. We propose a hierarchical
	"bottom up" cluster analysis based on unifying the elementary latent
	classes sequentially. The clustering procedure is controlled by minimum
	information loss criterion.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@INPROCEEDINGS{Grim2006,
  author = {Grim, J.},
  title = {EM cluster analysis for categorical data},
  booktitle = {Joint International Workshop on Structural, Syntactic, and Statistical
	Pattern Recognition},
  year = {2006},
  editor = {Yeung, D. Y. Kwok J. T. Fred A. Roli F. DeRidder D.},
  pages = {640-648},
  address = {Hong Kong, PEOPLES R CHINA},
  abstract = {Distribution mixtures with product components have been applied repeatedly
	to determine clusters in multivariate data. Unfortunately, for categorical
	variables the mixture parameters are not uniquely identifiable and
	therefore the result of cluster analysis may become questionable.
	We give a simple proof that any non-degenerate discrete product mixture
	can be equivalently described by infinitely many different parameter
	sets. Nevertheless a unique result of cluster analysis can be guaranteed
	by additional constraints. We propose a heuristic method of sequential
	estimation of components to guarantee a unique identification of
	clusters by means of EM algorithm. The application of the method
	is illustrated by a numerical example.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Gowda1992,
  author = {Gowda, K. C. and Diday, E.},
  title = {Symbolic clustering using a new similarity measure},
  journal = {Ieee Transactions on Systems Man and Cybernetics},
  year = {1992},
  volume = {22},
  pages = {368-378},
  number = {2},
  abstract = {A hierarchical, agglomerative, symbolic clustering methodology based
	on a new similarity measure that takes into consideration the "position,"
	"span," and "content," of symbolic objects is proposed. The similarity
	measure used is of a new type in the sense that it is not just another
	aspect of dissimilarity such as the reciprocal of a distance measure.
	The clustering methodology forms composite symbolic objects using
	a Cartesian join operator when two symbolic objects are merged. The
	maximum and minimum similarity values at various merging levels enable
	the determination of the number of clusters in the data set. The
	composite symbolic objects representing different clusters give a
	description of the resulting classes and lead to knowledge acquisition.
	The algorithm appears very versatile as it is capable of discerning
	clusters in data sets made up of numeric as well as symbolic objects
	consisting of different types and combinations of qualitative and
	quantitative feature values. In particular, the algorithm is applied
	on two data sets of fat-oil and microcomputers.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{McRae1971,
  author = {McRae, D. J.},
  title = {MIKCA - Fortran-IV iterative K-means cluster analysis program},
  journal = {Behavioral Science},
  year = {1971},
  volume = {16},
  pages = {423-\&},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{McLachlan2004,
  author = {McLachlan, G. J. and Chang, S. U.},
  title = {Mixture modelling for cluster analysis},
  journal = {Statistical Methods in Medical Research},
  year = {2004},
  volume = {13},
  pages = {347-361},
  number = {5},
  abstract = {Cluster analysis via a finite mixture model approach is considered.
	With this approach to clustering, the data can be partitioned into
	a specified number of clusters g by first fitting a mixture model
	with g components. An outright clustering of the data is then obtained
	by assigning an observation to the component to which it has the
	highest estimated posterior probability of belonging; that is, the
	ith cluster consists of those observations assigned to the ith component
	(i = 1,..., g). The focus is on the use of mixtures of normal components
	for the cluster analysis of data that can be regarded as being continuous.
	But attention is also given to the case of mixed data, where the
	observations consist of both continuous and discrete variables.},
  keywords = {EM ALGORITHM; LIKELIHOOD; ANALYZERS; CRITERIA; CHOICE},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@INPROCEEDINGS{Matthiessen2003,
  author = {Matthiessen, B. and Fock, H. O.},
  title = {A null model for the analysis of dietary overlap in Macroramphosus
	spp. at the Great Meteor Seamount (subtropical North-east Atlantic)},
  booktitle = {Census of Marine Life Symposium},
  year = {2003},
  pages = {294-304},
  address = {Newport, OR},
  abstract = {Macroramphosus spp. (Centriscidae, Macroramphosinae) are the most
	abundant demersal fishes on the Great Meteor Seamount (GMR, subtropical
	NE Atlantic, 30degrees N, 28.5degrees W). Previous investigations
	evidenced the existence of two species that differed in morphology
	and diet. A deep-bodied benthos feeding type (b-type), M. scolopax,
	could be significantly distinguished from a slender planktivorous
	type (p-type), M. gracilis. In this study we investigate whether
	dietary differences are also significant. Based on a novel null model
	termed ortho-RA3 in accordance with Lawlor (1980), the analysis of
	dietary overlap revealed significant differences between feeding
	types for two of three hypotheses that were tested, i.e., between-group
	differences and within-group differences for M. gracilis. Reasons
	for the failure of the third hypothesis (within-group differences
	for M scolopax) are discussed. Our results 4 reinforce the hypothesis
	of two competing species of snipefishes at GMR.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Marriott1982,
  author = {Marriott, F. H. C.},
  title = {Optimization methods of cluster analysis},
  journal = {Biometrika},
  year = {1982},
  volume = {69},
  pages = {417-421},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Marriott1971,
  author = {Marriott, F. H.},
  title = {Practical problems in a method of cluster analysis},
  journal = {Biometrics},
  year = {1971},
  volume = {27},
  pages = {501-\&},
  number = {3},
  note = {Times Cited: 73},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Maronna1974,
  author = {Maronna, R. and Jacovkis, P. M.},
  title = {Multivariate clustering procedures with variable metrics},
  journal = {Biometrics},
  year = {1974},
  volume = {30},
  pages = {499-505},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Manton2004,
  author = {Manton, K. G. and Gu, X. L. and Huang, H. and Kovtun, M.},
  title = {Fuzzy set analyses of genetic determinants of health and disability
	status},
  journal = {Statistical Methods in Medical Research},
  year = {2004},
  volume = {13},
  pages = {395-408},
  number = {5},
  abstract = {Analyses of complex genotype-phenotype relations require new statistical
	procedures because of the potentially high dimensionability of those
	relations which are expressed with both measurement error and stochasticity
	in the correlation function. We propose modifying a multivariate
	procedure called grade of membership (GoM) analysis to deal with
	the special problems of such analyses. In doing so, we make clear
	some special features of the GoM model for multivariate analysis
	of high dimensional, discrete data. This is illustrated for apolipoprotein
	E (APOE) assessments made on 1805 people in the 1999 National Long
	Term Care Survey. A number of interesting relations with APOE polymorphism
	were found where disability profiles were more predictive than specific
	diagnoses because they implicitly contained information on chronicity
	and severity of disease processes.},
  keywords = {APOLIPOPROTEIN-E POLYMORPHISM; CORONARY-ARTERY DISEASE; CELL-PROLIFERATION
	INDEXES; E PHENOTYPE; MYOCARDIAL-INFARCTION; HEART-DISEASE; E GENOTYPE;
	PLASMA-LIPOPROTEINS; RISK FACTOR; E ISOFORMS},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@MISC{Makoto/Takuji,
  author = {Makoto/Takuji},
  title = {Mersenne Twister RNG website},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Lunn1998,
  author = {Lunn, A. D. and Davies, S. J.},
  title = {A note on generating correlated binary variables},
  journal = {Biometrika},
  year = {1998},
  volume = {85},
  pages = {487-490},
  number = {2},
  abstract = {It is important to be able to generate correlated binary data in an
	efficient easily programmed manner for, among other things, the generation
	of large bootstrap samples. In this note nothing more than the basic
	ingredient of the uniform random number generator is required for
	simulating binary data from the most commonly occurring correlation
	structures.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Loganantharaj2006,
  author = {Loganantharaj, R. and Cheepala, S. and Clifford, J.},
  title = {Metric for measuring the effectiveness of clustering of DNA microarray
	expression},
  journal = {Bmc Bioinformatics},
  year = {2006},
  volume = {7},
  note = {3rd Annual Conference of the MidSouth-Computational-Biology-and-Bioformatics-Society
	MAR 02-04, 2006 Baton Rouge, LA},
  abstract = {Background: The recent advancement of microarray technology with lower
	noise and better affordability makes it possible to determine expression
	of several thousand genes simultaneously. The differentially expressed
	genes are filtered first and then clustered based on the expression
	profiles of the genes. A large number of clustering algorithms and
	distance measuring matrices are proposed in the literature. The popular
	ones among them include hierarchal clustering and k-means clustering.
	These algorithms have often used the Euclidian distance or Pearson
	correlation distance. The biologists or the practitioners are often
	confused as to which algorithm to use since there is no clear winner
	among algorithms or among distance measuring metrics. Several validation
	indices have been proposed in the literature and these are based
	directly or indirectly on distances; hence a method that uses any
	of these indices does not relate to any biological features such
	as biological processes or molecular functions. Results: In this
	paper we have proposed a metric to measure the effectiveness of clustering
	algorithms of genes by computing inter-cluster cohesiveness and as
	well as the intra-cluster separation with respect to biological features
	such as biological processes or molecular functions. We have applied
	this metric to the clusters on the data set that we have created
	as part of a larger study to determine the cancer suppressive mechanism
	of a class of chemicals called retinoids. We have considered hierarchal
	and k-means clustering with Euclidian and Pearson correlation distances.
	Our results show that genes of similar expression profiles are more
	likely to be closely related to biological processes than they are
	to molecular functions. The findings have been supported by many
	works in the area of gene clustering. Conclusion: The best clustering
	algorithm of genes must achieve cohesiveness within a cluster with
	respect to some biological features, and as well as maximum separation
	between clusters in terms of the distribution of genes of a behavioral
	group across clusters. We claim that our proposed metric is novel
	in this respect and that it provides a measure of both inter and
	intra cluster cohesiveness. Best of all, computation of the proposed
	metric is easy and it provides a single quantitative value, which
	makes comparison of different algorithms easier. The maximum cluster
	cohesiveness and the maximum intra-cluster separation are indicated
	by the metric when its value is 0. We have demonstrated the metric
	by applying it to a data set with gene behavioral groupings such
	as biological process and molecular functions. The metric can be
	easily extended to other features of a gene such as DNA binding sites
	and protein-protein interactions of the gene product, special features
	of the intron-exon structure, promoter characteristics, etc. The
	metric can also be used in other domains that use two different parametric
	spaces; one for clustering and the other one for measuring the effectiveness.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Gordon1994,
  author = {Gordon, A. D.},
  title = {Identifying genuine clusters in a classification},
  journal = {Computational Statistics \& Data Analysis},
  year = {1994},
  volume = {18},
  pages = {561-581},
  number = {5},
  abstract = {The paper addresses the problem of assessing the validity of clusters
	produced by a clustering procedure. Several null models for data
	are described. Previous research is reviewed, it being shown how
	much of it can be formulated in terms of properties of sets of within-cluster
	and between-cluster painwise dissimilarities. A Monte Carlo test
	for assessing the value of a U-statistic based on these sets of painwise
	dissimilarities is described and illustrated on four data sets. The
	final section includes further discussion of ways of specifying relevant
	null models.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Gange1995,
  author = {Gange, S. J.},
  title = {Generating multivariate categorical variates using the iterative
	proportional fitting algorithm},
  journal = {American Statistician},
  year = {1995},
  volume = {49},
  pages = {134-138},
  number = {2},
  abstract = {Two recent papers have suggested methods for generating correlated
	binary data with fixed marginal distributions and specified degrees
	of pairwise association. Emrich and Piedmonte suggested a method
	based on the existence of a multivariate normal distribution, while
	Lee suggested methods based on linear programming and Archimedian
	copulas. In this paper, a simpler method is described using the iterative
	proportional fitting algorithm for generating an n-dimensional distribution
	of correlated categorical data with specified margins of dimension
	1,2,..., k < n. An example of generating a distribution for a generalized
	estimating equations (GEE) model is discussed.},
  keywords = {CORRELATED OUTCOMES; GENERALIZED ESTIMATING EQUATIONS; LOG-LINEAR
	MODELS; RANDOM NUMBER GENERATION BINARY; ASSOCIATION; REGRESSION},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Fukunaga1970,
  author = {Fukunaga, K. and Koontz, W. L. G.},
  title = {A criterion and an algorithm for grouping data},
  journal = {Ieee Transactions on Computers},
  year = {1970},
  volume = {C 19},
  pages = {917-\&},
  number = {10},
  note = {Times Cited: 33},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Friedman1967,
  author = {Friedman, H. P. and Rubin, J.},
  title = {On some invariant criteria for grouping data},
  journal = {Journal of the American Statistical Association},
  year = {1967},
  volume = {62},
  pages = {1159-\&},
  number = {320},
  note = {Times Cited: 218},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Everitt1979,
  author = {Everitt, B. S.},
  title = {Unresolved problems in cluster analysis},
  journal = {Biometrics},
  year = {1979},
  volume = {35},
  pages = {169-181},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Edwards1965,
  author = {Edwards, A. W. F. and Cavallis.Ll},
  title = {A method for cluster analysis},
  journal = {Biometrics},
  year = {1965},
  volume = {21},
  pages = {362-\&},
  number = {2},
  note = {Times Cited: 160},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Dutta2005,
  author = {Dutta, M. and Mahanta, A. K. and Pujari, A. K.},
  title = {QROCK: A quick version of the ROCK algorithm for clustering of categorical
	data},
  journal = {Pattern Recognition Letters},
  year = {2005},
  volume = {26},
  pages = {2364-2373},
  number = {15},
  abstract = {The ROCK algorithm is an agglomerative hierarchical clustering algorithm
	for clustering categorical data [Guha S., Rastogi, R., Shim, K.,
	1999. ROCK: A robust clustering algorithm for categorical attributes.
	In: Proc. IEEE Internat. Conf. Data Engineering, Sydney, March 1999].
	In this paper we prove that under certain conditions, the final clusters
	obtained by the algorithm are nothing but the connected components
	of a certain graph with the input data-points as vertices. We propose
	a new algorithm QROCK which computes the clusters by determining
	the connected components of the graph. This leads to a very efficient
	method of obtaining the clusters giving a drastic reduction of the
	computing time of the ROCK algorithm. We also justify that it is
	more practical for specifying the similarity threshold rather than
	specifying the desired number of clusters a priori. The QROCK algorithm
	also detects the outliers in this process. We also discuss a new
	similarity measure for categorical attributes. (c) 2005 Elsevier
	B.V. All rights reserved.},
  keywords = {data clustering; categorical attributes; hierarchical clustering;
	similarity measure},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Dubes1987,
  author = {Dubes, R. C.},
  title = {How many clusters are best - an experiment},
  journal = {Pattern Recognition},
  year = {1987},
  volume = {20},
  pages = {645-663},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Dubes1979,
  author = {Dubes, R. and Jain, A. K.},
  title = {Validity studies in clustering methodologies},
  journal = {Pattern Recognition},
  year = {1979},
  volume = {11},
  pages = {235-254},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Dubes1976,
  author = {Dubes, R. and Jain, A. K.},
  title = {Clustering techniques - users dilemma},
  journal = {Pattern Recognition},
  year = {1976},
  volume = {8},
  pages = {247-260},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Leisch1998,
  author = {Leisch and Weingessel},
  title = {wp13 simulating binary data},
  year = {1998},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Lee1979,
  author = {Lee, K. L.},
  title = {Multivariate tests for clusters},
  journal = {Journal of the American Statistical Association},
  year = {1979},
  volume = {74},
  pages = {708-714},
  number = {367},
  note = {Times Cited: 35},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Kuncheva2006,
  author = {Kuncheva, L. I. and Vetrov, D. P.},
  title = {Evaluation of stability of k-means cluster ensembles with respect
	to random initialization},
  journal = {Ieee Transactions on Pattern Analysis and Machine Intelligence},
  year = {2006},
  volume = {28},
  pages = {1798-1808},
  number = {11},
  note = {Times Cited: 6},
  abstract = {Many clustering algorithms, including cluster ensembles, rely on a
	random component. Stability of the results across different runs
	is considered to be an asset of the algorithm. The cluster ensembles
	considered here are based on k-means clusterers. Each clusterer is
	assigned a random target number of clusters, k and is started from
	a random initialization. Here, we use 10 artificial and 10 real data
	sets to study ensemble stability with respect to random k, and random
	initialization. The data sets were chosen to have a small number
	of clusters (two to seven) and a moderate number of data points (up
	to a few hundred). Pairwise stability is defined as the adjusted
	Rand index between pairs of clusterers in the ensemble, averaged
	across all pairs. Nonpairwise stability is defined as the entropy
	of the consensus matrix of the ensemble. An experimental comparison
	with the stability of the standard k-means algorithm was carried
	out for k from 2 to 20. The results revealed that ensembles are generally
	more stable, markedly so for larger k. To establish whether stability
	can serve as a cluster validity index, we first looked at the relationship
	between stability and accuracy with respect to the number of clusters,
	k. We found that such a relationship strongly depends on the data
	set, varying from almost perfect positive correlation (0.97, for
	the glass data) to almost perfect negative correlation (-0.93, for
	the crabs data). We propose a new combined stability index to be
	the sum of the pairwise individual and ensemble stabilities. This
	index was found to correlate better with the ensemble accuracy. Following
	the hypothesis that a point of stability of a clustering algorithm
	corresponds to a structure found in the data, we used the stability
	measures to pick the number of clusters. The combined stability index
	gave best results.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Krzanowski1988,
  author = {Krzanowski, W. J. and Lai, Y. T.},
  title = {A criterion for determining the number of groups in a data set using
	sum-of-squares clustering},
  journal = {Biometrics},
  year = {1988},
  volume = {44},
  pages = {23-34},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Krummenauer1998,
  author = {Krummenauer, F.},
  title = {Efficient simulation of multivariate binomial and Poisson distributions},
  journal = {Biometrical Journal},
  year = {1998},
  volume = {40},
  pages = {823-832},
  number = {7},
  abstract = {Power investigations, for example, in statistical procedures for the
	assessment of agreement among multiple raters often require the simultaneous
	simulation of several dependent binomial or Poisson distributions
	to appropriately model the stochastical dependencies between the
	raters' results. Regarding the rather large dimensions of the random
	vectors to be generated and the even larger number of interactions
	to be introduced into the simulation scenarios to determine all necessary
	information on their distributions' dependence stucture, one needs
	efficient and fast algorithms for the simulation of multivariate
	Poisson and binomial distributions. Therefore two equivalent models
	for the multivariate Poisson distribution are combined to obtain
	an algorithm for the quick implementation of its multivariate dependence
	structure. Simulation of the multivariate Poisson distribution then
	becomes feasible by first generating and then convoluting independent
	univariate Poisson variates with appropriate expectations. The latter
	can be computed via linear recursion formulae. Similar means for
	simulation are also considered for the binomial setting. In this
	scenario it turns out, however, that exact computation of the probability
	function is even easier to perform; therefore corresponding linear
	recursion formulae for the point probabilities of multivariate binomial
	distributions are presented, which only require information about
	the index parameter and the (simultaneous) success probabilities,
	that is the multivariate dependence structure among the binomial
	marginals.},
  keywords = {multivariate discrete distributions; dependence structure; binomial
	distribution; Poisson distribution; simulation},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Krieger1999,
  author = {Krieger, A. M. and Green, P. E.},
  title = {A cautionary note on using internal cross validation to select the
	number of clusters},
  journal = {Psychometrika},
  year = {1999},
  volume = {64},
  pages = {341-353},
  number = {3},
  abstract = {A highly popular method for examining the stability of a data clustering
	is to split the data into two parts, cluster the observations in
	Part A, assign the objects in Part B to their nearest centroid in
	Part A, and then independently cluster the Part B objects. One then
	examines how close the two partitions are (say, by the Rand measure).
	Another proposal is to split the data into k parts, and see how their
	centroids cluster. By means of synthetic data analyses, we demonstrate
	that these approaches fail to identify the appropriate number of
	clusters, particularly as sample size becomes large and the variables
	exhibit higher correlations.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Kapp2007,
  author = {Kapp, A. V. and Tibshirani, R.},
  title = {Are clusters found in one dataset present in another dataset?},
  journal = {Biostatistics},
  year = {2007},
  volume = {8},
  pages = {9-31},
  number = {1},
  abstract = {In many microarray studies, a cluster defined on one dataset is sought
	in an independent dataset. If the cluster is found in the new dataset,
	the cluster is said to be "reproducible" and may be biologically
	significant. Classifying a new datum to a previously defined cluster
	can be seen as predicting which of the previously defined clusters
	is most similar to the new datum. If the new data classified to a
	cluster are similar, molecularly or clinically, to the data already
	present in the cluster, then the cluster is reproducible and the
	corresponding prediction accuracy is high. Here, we take advantage
	of the connection between reproducibility and prediction accuracy
	to develop a validation procedure for clusters found in datasets
	independent of the one in which they were characterized. We define
	a cluster quality measure called the "in-group proportion" (IGP)
	and introduce a general procedure for individually validating clusters.
	Using simulations and real breast cancer datasets, the IGP is compared
	to four other popular cluster quality measures (homogeneity score,
	separation score, silhouette width, and weighted average discrepant
	pairs score). Moreover, simulations and the real breast cancer datasets
	are used to compare the four versions of the validation procedure
	which all use the IGP, but differ in the way in which the null distributions
	are generated. We find that the IGP is the best measure of prediction
	accuracy, and one version of the validation procedure is the more
	widely applicable than the other three. An implementation of this
	algorithm is in a package called "clusterRepro" available through
	The Comprehensive R Archive Network (http://cran.r-project.org).},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@INPROCEEDINGS{Juan2004a,
  author = {Juan, A. and Vidal, E.},
  title = {Bernoulli mixture models for binary images},
  booktitle = {17th International Conference on Pattern Recognition (ICPR)},
  year = {2004},
  editor = {Kittler, J. Petrou M. Nixon M.},
  pages = {367-370},
  address = {Cambridge, ENGLAND},
  abstract = {Mixture modelling is a hot area in pattern recognition. Although most
	research in this area has focused on mixtures for continuous data,
	there are many pattern recognition tasks for which binary or discrete
	mixtures are better suited. This paper focuses on the use of Bernoulli
	mixtures for binary data and, in particular for binary images. Results
	are reported on a task of handwritten Indian digits.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@INPROCEEDINGS{Juan2001,
  author = {Juan, A. and Vidal, E.},
  title = {On the use of Bernoulli mixture models for text classification},
  booktitle = {1st International Workshop on Pattern Recognition in Information
	Systems},
  year = {2001},
  pages = {2705-2710},
  address = {Setubal, Portugal},
  abstract = {Mixture modelling of class-conditional densities is a standard pattern
	recognition technique. Although most research on mixture models has
	concentrated on mixtures for continuous data, emerging pattern recognition
	applications demand extending research efforts to other data types.
	This paper focuses on the application of mixtures of multivariate
	Bernoulli distributions to binary data. More concretely, a text classification
	task aimed at improving language modelling for machine translation
	is considered. (C) 2002 Pattern Recognition Society. Published by
	Elsevier Science Ltd. All rights reserved.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Yin2008,
  author = {Yin, Z. and Zhou, X. B. and Bakal, C. and Li, F. H. and Sun, Y. X.
	and Perrimon, N. and Wong, S. T.},
  title = {Using iterative cluster merging with improved gap statistics to perform
	online phenotype discovery in the context of high-throughput RNAi
	screens},
  journal = {Bmc Bioinformatics},
  year = {2008},
  volume = {9},
  abstract = {Background: The recent emergence of high-throughput automated image
	acquisition technologies has forever changed how cell biologists
	collect and analyze data. Historically, the interpretation of cellular
	phenotypes in different experimental conditions has been dependent
	upon the expert opinions of well-trained biologists. Such qualitative
	analysis is particularly effective in detecting subtle, but important,
	deviations in phenotypes. However, while the rapid and continuing
	development of automated microscope-based technologies now facilitates
	the acquisition of trillions of cells in thousands of diverse experimental
	conditions, such as in the context of RNA interference (RNAi) or
	small-molecule screens, the massive size of these datasets precludes
	human analysis. Thus, the development of automated methods which
	aim to identify novel and biological relevant phenotypes online is
	one of the major challenges in high-throughput image-based screening.
	Ideally, phenotype discovery methods should be designed to utilize
	prior/existing information and tackle three challenging tasks, i.e.
	restoring pre-defined biological meaningful phenotypes, differentiating
	novel phenotypes from known ones and clarifying novel phenotypes
	from each other. Arbitrarily extracted information causes biased
	analysis, while combining the complete existing datasets with each
	new image is intractable in high-throughput screens. Results: Here
	we present the design and implementation of a novel and robust online
	phenotype discovery method with broad applicability that can be used
	in diverse experimental contexts, especially high-throughput RNAi
	screens. This method features phenotype modelling and iterative cluster
	merging using improved gap statistics. A Gaussian Mixture Model (GMM)
	is employed to estimate the distribution of each existing phenotype,
	and then used as reference distribution in gap statistics. This method
	is broadly applicable to a number of different types of image-based
	datasets derived from a wide spectrum of experimental conditions
	and is suitable to adaptively process new images which are continuously
	added to existing datasets. Validations were carried out on different
	dataset, including published RNAi screening using Drosophila embryos
	[ Additional files 1, 2], dataset for cell cycle phase identification
	using HeLa cells [ Additional files 1, 3, 4] and synthetic dataset
	using polygons, our methods tackled three aforementioned tasks effectively
	with an accuracy range of 85%-90%. When our method is implemented
	in the context of a Drosophila genome-scale RNAi image-based screening
	of cultured cells aimed to identifying the contribution of individual
	genes towards the regulation of cell-shape, it efficiently discovers
	meaningful new phenotypes and provides novel biological insight.
	We also propose a two-step procedure to modify the novelty detection
	method based on one-class SVM, so that it can be used to online phenotype
	discovery. In different conditions, we compared the SVM based method
	with our method using various datasets and our methods consistently
	outperformed SVM based method in at least two of three tasks by 2%
	to 5%. These results demonstrate that our methods can be used to
	better identify novel phenotypes in image-based datasets from a wide
	range of conditions and organisms. Conclusion: We demonstrate that
	our method can detect various novel phenotypes effectively in complex
	datasets. Experiment results also validate that our method performs
	consistently under different order of image input, variation of starting
	conditions including the number and composition of existing phenotypes,
	and dataset from different screens. In our findings, the proposed
	method is suitable for online phenotype discovery in diverse high-throughput
	image-based genetic and chemical screens.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@INPROCEEDINGS{Juan2004,
  author = {Juan, A. and Garcia-Hernandez, J. and Vidal, E.},
  title = {EM initialisation for Bernoulli mixture learning},
  booktitle = {10th International Workshop on Structural and Syntactic Pattern Recognition/5th
	International Conference on Statistical Techniques in Pattern Recognition},
  year = {2004},
  editor = {Fred, A. Caelli T. Duin R. P. W. Campilho A. DeRidder D.},
  pages = {635-643},
  address = {Lisbon, PORTUGAL},
  abstract = {Mixture modelling is a hot area in pattern recognition. This paper
	focuses on the use of Bernoulli mixtures for binary data and, in
	particular, for binary images. More specifically, six EM initialisation
	techniques are described and empirically compared on a classification
	task of handwritten Indian digits. Somehow surprisingly, we have
	found that a relatively good initialisation for Bernoulli prototypes
	is to use slightly perturbed versions of the hypercube centre.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Yeung2001a,
  author = {Yeung, K. Y. and Ruzzo, W. L.},
  title = {Principal component analysis for clustering gene expression data},
  journal = {Bioinformatics},
  year = {2001},
  volume = {17},
  pages = {763-774},
  number = {9},
  abstract = {Motivation: There is a great need to develop analytical methodology
	to analyze and to exploit the information contained in gene expression
	data. Because of the large number of genes and the complexity of
	biological networks, clustering is a useful exploratory technique
	for analysis of gene expression data. Other classical techniques,
	such as principal component analysis (PCA), have also been applied
	to analyze gene expression data. Using different data analysis techniques
	and different clustering algorithms to analyze the same data set
	can lead to very different conclusions. Our goal is to study the
	effectiveness of principal components (PCs) in capturing cluster
	structure. Specifically, using both real and synthetic gene expression
	data sets, we compared the quality of clusters obtained from the
	original data to the quality of clusters obtained after projecting
	onto subsets of the principal component axes. Results: Our empirical
	study showed that clustering with the PCs instead of the original
	variables does not necessarily improve, and often degrades, cluster
	quality. In particular, the first few PCs (which contain most of
	the variation in the data) do not necessarily capture most of the
	cluster structure. We also showed that clustering with PCs has different
	impact on different algorithms and different similarity metrics.
	Overall, we would not recommend PCA before clustering except In special
	circumstances.},
  keywords = {PATTERNS; CRITERIA; ARRAY},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Yeung2001,
  author = {Yeung, K. Y. and Haynor, D. R. and Ruzzo, W. L.},
  title = {Validating clustering for gene expression data},
  journal = {Bioinformatics},
  year = {2001},
  volume = {17},
  pages = {309-318},
  number = {4},
  abstract = {Motivation: Many clustering algorithms have been proposed for the
	analysis of gene expression data, but little guidance is available
	to help choose among them. We provide a systematic framework for
	assessing the results of clustering algorithms. Clustering algorithms
	attempt to partition the genes into groups exhibiting similar patterns
	of variation in expression level. Our methodology is to apply a clustering
	algorithm to the data from all but one experimental condition. The
	remaining condition is used to assess the predictive power of the
	resulting clusters-meaningful clusters should exhibit less variation
	in the remaining condition than clusters formed by chance. Results:
	We successfully applied our methodology to compare six clustering
	algorithms on four gene expression data sets. We found our quantitative
	measures of cluster quality to be positively correlated with external
	standards of cluster quality.},
  keywords = {PATTERNS; NUMBER},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Dolcinar1998,
  author = {Dolcinar},
  title = {WP19- more comparison},
  year = {1998},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Windham1987,
  author = {Windham, M. P.},
  title = {Parameter modification for clustering criteria},
  journal = {Journal of Classification},
  year = {1987},
  volume = {4},
  pages = {191-214},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Dolcinar1998a,
  author = {Dolcinar},
  title = {WP7: comparison of clustering algorithms},
  year = {1998},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Waller1999,
  author = {Waller, N. G. and Underhill, J. M. and Kaiser, H. A.},
  title = {A method for generating simulated plasmodes and artificial test clusters
	with user-defined shape, size, and orientation},
  journal = {Multivariate Behavioral Research},
  year = {1999},
  volume = {34},
  pages = {123-142},
  number = {2},
  abstract = {We present a simple method for generating simulated plasmodes and
	artificial test clusters with user-defined shape, size, and orientation.
	Our method differs from other cluster generation techniques in that
	it focuses on the validity of the duster indicators. For J clusters,
	indicator validity is defined as the squared correlation ratio between
	the cluster indicator (i.e., the observed variable) and J-1 dummy
	variables. The within-cluster correlation structure and the univariate
	distributions of the cluster indicators are specified with procedures
	outlined by Fleishman (1978) and Vale and Maurelli (1983). Simulation
	results illustrate the utility of the method for cluster analysis
	evaluation research.},
  keywords = {MIXTURE MODEL TESTS; ANALYSIS ALGORITHMS; ACCURACY},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Dimitriadou2002,
  author = {Dimitriadou, E. and Dolnicar, S. and Weingessel, A.},
  title = {An examination of indexes for determining the number of clusters
	in binary data sets},
  journal = {Psychometrika},
  year = {2002},
  volume = {67},
  pages = {137-159},
  number = {1},
  abstract = {The problem of choosing the correct number of clusters is as old as
	cluster analysis itself. A number of authors have suggested various
	indexes to facilitate this crucial decision. One of the most extensive
	comparative studies of indexes was conducted by Milligan and Cooper
	(1985). The present piece of work pursues the same goal under different
	conditions. In contrast to Milligan and Cooper's work, the emphasis
	here is on high-dimensional empirical binary data. Binary artificial
	data sets are constructed to reflect features typically encountered
	in real-world data situations in the field of marketing research.
	The simulation includes 162 binary data sets that are clustered by
	two different algorithms and lead to recommendations on the number
	of clusters for each index under consideration. Index results are
	evaluated and their performance is compared and analyzed.},
  keywords = {number of clusters; clustering indexes; binary data; artificial data
	sets; market segmentation SIMILARITY; COEFFICIENTS},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Vassiliou1989,
  author = {Vassiliou, A. and Ignatiades, L. and Karydis, M.},
  title = {Clustering of transect phytoplankton collections with a quick randomization
	algorithm},
  journal = {Journal of Experimental Marine Biology and Ecology},
  year = {1989},
  volume = {130},
  pages = {135-145},
  number = {2},
  note = {Times Cited: 9},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Datta2003,
  author = {Datta, S. and Datta, S.},
  title = {Comparisons and validation of statistical clustering techniques for
	microarray gene expression data},
  journal = {Bioinformatics},
  year = {2003},
  volume = {19},
  pages = {459-466},
  number = {4},
  abstract = {Motivation: With the advent of microarray chip technology, large data
	sets are emerging containing the simultaneous expression levels of
	thousands of genes at various time points during a biological process.
	Biologists are attempting to group genes based on the temporal pattern
	of their expression levels. While the use of hierarchical clustering
	(UPGMA) with correlation 'distance' has been the most common in the
	microarray studies, there are many more choices of clustering algorithms
	in pattern recognition and statistics literature. At the moment there
	do not seem to be any clear-cut guidelines regarding the choice of
	a clustering algorithm to be used for grouping genes based on their
	expression profiles. Results: In this paper, we consider six clustering
	algorithms (of various flavors!) and evaluate their performances
	on a well-known publicly available microarray data set on sporulation
	of budding yeast and on two simulated data sets. Among other things,
	we formulate three reasonable validation strategies that can be used
	with any clustering algorithm when temporal observations or replications
	are present. We evaluate each of these six clustering methods with
	these validation measures. While the 'best' method is dependent on
	the exact validation strategy and the number of clusters to be used,
	overall Diana appears to be a solid performer. Interestingly, the
	performance of correlation-based hierarchical clustering and model-based
	clustering (another method that has been advocated by a number of
	researchers) appear to be on opposite extremes, depending on what
	validation measure one employs. Next it is shown that the group means
	produced by Diana are the closest and those produced by UPGMA are
	the farthest from a model profile based on a set of hand-picked genes.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Chaganty2006,
  author = {Chaganty, N. R. and Joe, H.},
  title = {Range of correlation matrices for dependent Bernoulli random variables},
  journal = {Biometrika},
  year = {2006},
  volume = {93},
  pages = {197-206},
  number = {1},
  abstract = {We say that a pair (p, R) is compatible if there exists a multivariate
	binary distribution with mean vector p and correlation matrix R.
	In this paper we study necessary and sufficient conditions for compatibility
	for structured and unstructured correlation matrices. We give examples
	of correlation matrices that are incompatible with any p. Using our
	results we show that the parametric binary models of Emrich & Piedmonte
	(1991) and Qaqish (2003) allow a good range of correlations between
	the binary variables. We also obtain necessary and sufficient conditions
	for a matrix of odds ratios to be compatible with a given p. Our
	findings support the popular belief that the odds ratios are less
	constrained and more flexible than the correlations.},
  keywords = {Frechet bound; generalised estimating equation; multivariate binary;
	odds ratio BINARY VARIABLES; DISTRIBUTIONS},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Cesario2007,
  author = {Cesario, E. and Manco, G. and Ortale, R.},
  title = {Top-down parameter-free clustering of high-dimensional categorical
	data},
  journal = {Ieee Transactions on Knowledge and Data Engineering},
  year = {2007},
  volume = {19},
  pages = {1607-1624},
  number = {12},
  note = {Times Cited: 0},
  abstract = {A parameter-free, fully-automatic approach to clustering high-dimensional
	categorical data is proposed. The technique is based on a two-phase
	iterative procedure, which attempts to improve the overall quality
	of the whole partition. In the first phase, cluster assignments are
	given, and a new cluster is added to the partition by identifying
	and splitting a low-quality cluster. In the second phase, the number
	of clusters is fixed, and an attempt to optimize cluster assignments
	is done. On the basis of such features, the algorithm attempts to
	improve the overall quality of the whole partition and finds clusters
	in the data, whose number is naturally established on the basis of
	the inherent features of the underlying data set rather than being
	previously specified. Furthermore, the approach is parametric to
	the notion of cluster quality: Here, a cluster is defined as a set
	of tuples exhibiting a sort of homogeneity. We show how a suitable
	notion of cluster homogeneity can be defined in the context of high-dimensional
	categorical data, from which an effective instance of the proposed
	clustering scheme immediately follows. Experiments on both synthetic
	and real data prove that the devised algorithm scales linearly and
	achieves nearly optimal results in terms of compactness and separation.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Cary,
  author = {Cary, A. J.},
  title = {Generating data with the SAS dataset},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Carreira-Perpinan2000,
  author = {Carreira-Perpinan, M. A. and Renals, S.},
  title = {Practical identifiability of finite mixtures of multivariate Bernoulli
	distributions},
  journal = {Neural Computation},
  year = {2000},
  volume = {12},
  pages = {141-152},
  number = {1},
  note = {Times Cited: 9},
  abstract = {The class of finite mixtures of multivariate Bernoulli distributions
	is known to be nonidentifiable; that is, different values of the
	mixture parameters can correspond to exactly the same probability
	distribution. In principle, this would mean that sample estimates
	using this model would give rise to different interpretations. We
	give empirical support to the fact that estimation of this class
	of mixtures can still produce meaningful results in practice, thus
	lessening the importance of the identifiability problem. We also
	show that the expectation-maximization algorithm is guaranteed to
	converge to a proper maximum likelihood estimate, owing to a property
	of the log-likelihood surface. Experiments with synthetic data sets
	show that an original generating distribution can be estimated from
	a sample. Experiments with an electropalatography data set show important
	structure in the data.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Calinski1968,
  author = {Calinski, T.},
  title = {A dendrite method for cluster analysis},
  journal = {Biometrics},
  year = {1968},
  volume = {24},
  pages = {207-\&},
  number = {1},
  note = {Times Cited: 2},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Bryant1991,
  author = {Bryant, P. G.},
  title = {Large-sample results for optimization-based clustering methods},
  journal = {Journal of Classification},
  year = {1991},
  volume = {8},
  pages = {31-44},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.20}
}

@ARTICLE{Pickles1995,
  author = {Pickles, A. and Bolton, P. and Macdonald, H. and Bailey, A. and Lecouteur,
	A. and Sim, C. H. and Rutter, M.},
  title = {Latent-class analysis of recurrence risks for complex phenotypes
	with selection and measurement error - a twin and family history
	study of autism},
  journal = {American Journal of Human Genetics},
  year = {1995},
  volume = {57},
  pages = {717-726},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.21}
}

@ARTICLE{McLachlan1987,
  author = {McLachlan, G. J.},
  title = {On bootstrapping the likelihood ratio test statistic for the number
	of components in a normal mixture},
  journal = {Applied Statistics-Journal of the Royal Statistical Society Series
	C},
  year = {1987},
  volume = {36},
  pages = {318-324},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.21}
}

@ARTICLE{Loken2004,
  author = {Loken, E.},
  title = {Using latent class analysis to model temperament types},
  journal = {Multivariate Behavioral Research},
  year = {2004},
  volume = {39},
  pages = {625-652},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.21}
}

@ARTICLE{Lanza2007,
  author = {Lanza, S. T. and Collins, L. M. and Lemmon, D. R. and Schafer, J.
	L.},
  title = {PROC LCA: A SAS procedure for latent class analysis},
  journal = {Structural Equation Modeling-a Multidisciplinary Journal},
  year = {2007},
  volume = {14},
  pages = {671-694},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.21}
}

@ARTICLE{Holt1989,
  author = {Holt, J. A. and Macready, G. B.},
  title = {A simulation study of the difference Chi-square statistic for comparing
	latent class models under violation of regularity conditions},
  journal = {Applied Psychological Measurement},
  year = {1989},
  volume = {13},
  pages = {221-231},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.21}
}

@ARTICLE{Feng1996,
  author = {Feng, Z. D. and McCulloch, C. E.},
  title = {Using bootstrap likelihood ratios in finite mixture models},
  journal = {Journal of the Royal Statistical Society Series B-Methodological},
  year = {1996},
  volume = {58},
  pages = {609-617},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.21}
}

@ARTICLE{Everitt1988,
  author = {Everitt, B. S.},
  title = {A Monte-Carlo investigation of the likelihood ratio test for number
	of classes in latent class analysis},
  journal = {Multivariate Behavioral Research},
  year = {1988},
  volume = {23},
  pages = {531-538},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.21}
}

@ARTICLE{Desoete1991,
  author = {Desoete, G. and Desarbo, W. S.},
  title = {A latent class probit model for analyzing pick any/n data},
  journal = {Journal of Classification},
  year = {1991},
  volume = {8},
  pages = {45-63},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.21}
}

@ARTICLE{Banfield1993,
  author = {Banfield, J. D. and Raftery, A. E.},
  title = {Model-based gaussian and non-gaussian clustering},
  journal = {Biometrics},
  year = {1993},
  volume = {49},
  pages = {803-821},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.21}
}

@ARTICLE{Aitkin1981,
  author = {Aitkin, M. and Anderson, D. and Hinde, J.},
  title = {Statistical modeling of data on teaching styles},
  journal = {Journal of the Royal Statistical Society Series a-Statistics in Society},
  year = {1981},
  volume = {144},
  pages = {419-461},
  note = {Part 4},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.21}
}

@BOOK{Krzanowski,
  title = {Multivariate Analysis Part 2: Classification, covariance structures
	and repeated measurements},
  publisher = {Halsted Press},
  author = {Krzanowski, W. J. and Marriott, F. H. C.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.22}
}

@ARTICLE{Garrett2000,
  author = {Garrett, E. S. and Zeger, S. L.},
  title = {Latent class model diagnosis},
  journal = {Biometrics},
  year = {2000},
  volume = {56},
  pages = {1055-1067},
  number = {4},
  abstract = {In many areas of medical research, such as psychiatry and gerontology,
	latent class variables are used to classify individuals into disease
	categories, often with the intention of hierarchical modeling. Problems
	arise when it is not clear how many disease classes are appropriate,
	creating a need for model selection and diagnostic techniques. Previous
	work has shown that the Pearson chi (2) statistic and the log-likelihood
	ratio G(2) statistic are not valid test statistics for evaluating
	latent class models. Other methods, such as information criteria,
	provide decision rules without providing explicit information about
	where discrepancies occur between a model and the data. Identifiability
	issues further complicate these problems. This paper develops procedures
	for assessing Markov chain Monte Carlo convergence and model diagnosis
	and for selecting the number of categories for the latent variable
	based on evidence in the data using Markov chain Monte Carlo techniques.
	Simulations and a psychiatric example are presented to demonstrate
	the effective use of these methods.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.23}
}

@ARTICLE{Garrett2002,
  author = {Garrett, E. S. and Eaton, W. W. and Zeger, S.},
  title = {Methods for evaluating the performance of diagnostic tests in the
	absence of a gold standard: a latent class model approach},
  journal = {Statistics in Medicine},
  year = {2002},
  volume = {21},
  pages = {1289-1307},
  number = {9},
  abstract = {In many areas of medical research, `gold standard' diagnostic tests
	do not exist and so evaluating the performance of standardized diagnostic
	criteria or algorithms is problematic. In this paper we propose an
	approach to evaluating the operating characteristics of diagnoses
	using a latent class model. By defining `true disease' as our latent
	variable, we are able to estimate sensitivity, specificity and negative
	and positive predictive values of the diagnostic test. These methods
	are applied to diagnostic criteria for depression using Baltimore's
	Epidemiologic Catchment Area Study Wave 3 data. Copyright (C) 2002
	John Wiley Sons, Ltd.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.23}
}

@ARTICLE{Zeger2003,
  author = {Zeger, S. L. and Garrett, E. S.},
  title = {Rejoinder to "Latent class model diagnosis from a frequentist point
	of view"},
  journal = {Biometrics},
  year = {2003},
  volume = {59},
  pages = {197-198},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.23}
}

@ARTICLE{Uebersax1999,
  author = {Uebersax, J. S.},
  title = {Probit latent class analysis with dichotomous or ordered category
	measures: Conditional independence/dependence models},
  journal = {Applied Psychological Measurement},
  year = {1999},
  volume = {23},
  pages = {283-297},
  number = {4},
  abstract = {Flexible methods that relax restrictive conditional independence assumptions
	of latent class analysis (LCA) are described. Dichotomous and ordered
	category manifest variables are viewed as discretized latent continuous
	variables. The latent continuous variables are assumed to have a
	mixture-of-multivariate-normals distribution. Within a latent class,
	conditional dependence is modeled as the mutual association of all
	or some latent continuous variables with a continuous latent trait
	(or in special cases, multiple latent traits). The relaxation of
	conditional independence assumptions allows LCA to better model natural
	taxa. Comparisons of specific restricted and unrestricted models
	permit statistical tests of specific aspects of latent taxonic structure.
	Latent class, latent trait, and latent distribution analysis can
	be viewed as special cases of the mixed latent trait model. The relationship
	between the multivariate probit mixture model proposed here and Rest's
	mixed Rasch(1990, 1991) model is discussed. Two studies illustrate
	different uses of the proposed model.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.23}
}

@ARTICLE{Reboussin2008,
  author = {Reboussin, B. A. and Ip, E. H. and Wolfson, M.},
  title = {Locally dependent latent class models with covariates: an application
	to under-age drinking in the USA},
  journal = {Journal of the Royal Statistical Society Series a-Statistics in Society},
  year = {2008},
  volume = {171},
  pages = {877-897},
  note = {Part 4},
  abstract = {Under-age drinking is a long-standing public health problem in the
	USA and the identification of underage drinkers suffering alcohol-related
	problems has been difficult by using diagnostic criteria that were
	developed in adult populations. For this reason, it is important
	to characterize patterns of drinking in adolescents that are associated
	with alcohol-related problems. Latent class analysis is a statistical
	technique for explaining heterogeneity in individual response patterns
	in terms of a smaller number of classes. However, the latent class
	analysis assumption of local independence may not be appropriate
	when examining behavioural profiles and could have implications for
	statistical inference. In addition, if covariates are included in
	the model, non-differential measurement is also assumed. We propose
	a flexible set of models for local dependence and differential measurement
	that use easily interpretable odds ratio parameterizations while
	simultaneously fitting a marginal regression model for the latent
	class prevalences. Estimation is based on solving a set of second-order
	estimating equations. This approach requires only specification of
	the first two moments and allows for the choice of simple 'working'
	covariance structures. The method is illustrated by using data from
	a large-scale survey of under-age drinking. This new approach indicates
	the effectiveness of introducing local dependence and differential
	measurement into latent class models for selecting substantively
	interpretable models over more complex models that are deemed empirically
	superior.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.23}
}

@ARTICLE{Menten2008,
  author = {Menten, J. and Boelaert, M. and Lesaffre, E.},
  title = {Bayesian latent class models with conditionally dependent diagnostic
	tests: A case study},
  journal = {Statistics in Medicine},
  year = {2008},
  volume = {27},
  pages = {4469-4488},
  number = {22},
  abstract = {In the assessment of the accuracy of diagnostic tests for infectious
	diseases, the true disease status of the subjects is often unknown
	due to the lack of a gold standard test. Latent class models with
	two latent classes, representing diseased and non-diseased subjects,
	are often used to analyze this type of data. In its basic format,
	latent class analysis requires the observed outcomes to be statistically
	independent conditional on the disease status. In most diagnostic
	settings, this assumption is highly questionable. During the last
	decade, several methods have been proposed to estimate latent class
	models with conditional dependence between the test results. A class
	of flexible fixed and random effects models were described by Dendukuri
	and Joseph in a Bayesian framework. We illustrate these models using
	the analysis of a diagnostic study of three field tests and an imperfect
	reference test for the diagnosis of visceral leishmaniasis. We show
	that, as observed earlier by Albert and Dodd, different dependence
	models may result in similar fits to the data while resulting in
	different inferences. Given this problem, selection of appropriate
	latent class models should be based on substantive subject matter
	knowledge. If several clinically plausible models are supported by
	the data, a sensitivity analysis should be performed by describing
	the results obtained from different models and using different priors.
	Copyright (c) 2008 John Wiley & ons, Ltd.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.23}
}

@ARTICLE{Branscum2004,
  author = {Branscum, A. J. and Gardner, I. A. and Johnson, W. O.},
  title = {Bayesian modeling of animal- and herd-level prevalences},
  journal = {Preventive Veterinary Medicine},
  year = {2004},
  volume = {66},
  pages = {101-112},
  number = {1-4},
  abstract = {We reviewed Bayesian approaches for animal-level and herd-level prevalence
	estimation based on cross-sectional sampling designs and demonstrated
	fitting of these models using the WinBUGS software. We considered
	estimation of infection prevalence based on use of a single diagnostic
	test applied to a single herd with binomial and hypergeometric sampling.
	We then considered multiple herds under binomial sampling with the
	primary goal of estimating the prevalence distribution and the proportion
	of infected herds. A new model is presented that can be used to estimate
	the herd-level prevalence in a region, including the posterior probability
	that all herds are non-infected. Using this model, inferences for
	the distribution of prevalences, mean prevalence in the region, and
	predicted prevalence of herds in the region (including the predicted
	probability of zero prevalence) are also available. In the models
	presented, both animal- and herd-level prevalences are modeled as
	mixture distributions to allow for zero infection prevalences. (If
	mixture models for the prevalences were not used, prevalence estimates
	might be artificially inflated, especially in herds and regions with
	low or zero prevalence.) Finally, we considered estimation of animal-level
	prevalence based on pooled samples. (C) 2004 Elsevier B.V. All rights
	reserved.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.23}
}

@BOOK{Everitt1981,
  title = {Finite mixture distributions},
  publisher = {Chapman and Hall},
  year = {1981},
  author = {Everitt, Brian and Hand, D. J.},
  series = {Monographs on applied probability and statistics},
  address = {London ; New York},
  keywords = {Mixture distributions (Probability theory)},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.24}
}

@BOOK{Everitt1993,
  title = {Cluster analysis},
  publisher = {E. Arnold ; Halsted Press},
  year = {1993},
  author = {Everitt, Brian},
  address = {London New York},
  edition = {3rd},
  keywords = {Cluster analysis.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.24}
}

@BOOK{Everitt1984,
  title = {An introduction to latent variable models},
  publisher = {Chapman and Hall},
  year = {1984},
  author = {Everitt, Brian},
  series = {Monographs on statistics and applied probability},
  address = {London ; New York},
  keywords = {Latent variables.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.24}
}

@BOOK{Krzanowski1994,
  title = {Multivariate analysis},
  publisher = {E. Arnold ; Halsted Press},
  year = {1994},
  author = {Krzanowski, W. J. and Marriott, F. H. C.},
  series = {Kendall's library of statistics},
  address = {London New York},
  keywords = {Multivariate analysis.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.24}
}

@BOOK{Everitt1974,
  title = {Cluster analysis},
  publisher = {Heinemann Educational [for] the Social Science Research Council},
  year = {1974},
  author = {Everitt, Brian},
  address = {London},
  keywords = {Cluster analysis.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.24}
}

@BOOK{Dunn1993,
  title = {Modelling covariances and latent variables using EQS},
  publisher = {Chapman \& Hall},
  year = {1993},
  author = {Dunn, G. and Everitt, Brian and Pickles, Andrew},
  address = {London ; New York},
  edition = {1st},
  note = {93244250 G. Dunn, B. Everitt, and A. Pickles. ill. ; 24 cm. Reference
	card inserted. Includes bibliographical references (p. [193]-194)
	and index.},
  keywords = {Factor analysis Data processing. Social sciences Statistical methods
	Data processing.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.24}
}

@BOOK{Everitt1980,
  title = {Cluster analysis},
  publisher = {published on behalf of the Social Science Research Council by Heinemann
	Educational Books ; Halsted Press},
  year = {1980},
  author = {Everitt, Brian},
  address = {London New York},
  edition = {2d},
  keywords = {Cluster analysis.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.24}
}

@BOOK{Everitt2001,
  title = {Cluster analysis},
  publisher = {Arnold ; Oxford University Press},
  year = {2001},
  author = {Everitt, Brian and Landau, Sabine and Leese, Morven},
  address = {London New York},
  edition = {4th},
  keywords = {Cluster analysis.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.24}
}

@ARTICLE{Escobar1995,
  author = {Escobar, M. D. and West, M.},
  title = {Bayesian density estimation and inference using mixtures},
  journal = {Journal of the American Statistical Association},
  year = {1995},
  volume = {90},
  pages = {577-588},
  number = {430},
  abstract = {We describe and illustrate Bayesian inference in models for density
	estimation using mixtures of Dirichlet processes. These models provide
	natural settings for density estimation and are exemplified by special
	eases where data are modeled as a sample from mixtures of normal
	distributions. Efficient simulation methods are used to approximate
	various prior, posterior, and predictive distributions. This allows
	for direct inference on a variety of practical issues, including
	problems of local versus global smoothing, uncertainty about density
	estimates, assessment of modality, and the inference on the numbers
	of components. Also, convergence results are established for a general
	class of normal mixture models.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Eaton1989,
  author = {Eaton, W. W. and Dryman, A. and Sorenson, A. and McCutcheon, A.},
  title = {DSM-III major depressive disorder in the community: a latent class
	analysis of data from the NIMH Epidemiologic Catchment Area Program},
  journal = {British Journal of Psychiatry},
  year = {1989},
  volume = {155},
  pages = {48-54},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Dayton1988,
  author = {Dayton, C. M. and Macready, G. B.},
  title = {Concomitant-variable latent-class models},
  journal = {Journal of the American Statistical Association},
  year = {1988},
  volume = {83},
  pages = {173-178},
  number = {401},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@INPROCEEDINGS{Collins1992,
  author = {Collins, L. M. and Fidler, P. L. and Wugalter, S. E. and Long, J.
	D.},
  title = {Goodness-of-fit testing for latent class models},
  booktitle = {1992 Annual Meeting of the Soc-of-Multivariate-Experimental-Psychology},
  year = {1992},
  pages = {375-389},
  address = {Chatham, Ma},
  abstract = {Latent class models with sparse contingency tables can present problems
	for model comparison and selection, because under these conditions
	the distributions of goodness-of-fit indices are often unknown. This
	causes inaccuracies both in hypothesis testing and in model comparisons
	based on normed indices. In order to assess the extent of this problem,
	we carried out a simulation investigating the distributions of the
	likelihood ratio statistic G2, the Pearson statistic X2, and a new
	goodness-of-fit index suggested by Read and Cressie (1988). There
	were substantial deviations between the expectation of the chi-squared
	distribution and the means of the G2 and Read and Cressie distributions.
	In general, the mean of the distribution of a statistic was closer
	to the expectation of the chi-squared distribution when the average
	cell expectation was large. there were fewer indicator items, and
	the latent class measurement parameters were less extreme. It was
	found that the mean of the X2 distribution is generally closer to
	the expectation of the chi-squared distribution than are the means
	of the other two indices we examined, but the standard deviation
	of the X2 distribution is considerably larger than that of the other
	two indices and larger than the standard deviation of the chi-squared
	distribution. We argue that a possible solution is to forgo reliance
	on theoretical distributions for expectations and quantiles of goodness-of-fit
	statistics. Instead, Monte Carlo sampling (Noreen, 1989) can be used
	to arrive at an empirical central or noncentral distribution.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Chib1995,
  author = {Chib, S. and Greenberg, E.},
  title = {Understanding the Metropolis-Hastings algorithms},
  journal = {American Statistician},
  year = {1995},
  volume = {49},
  pages = {327-335},
  number = {4},
  abstract = {We provide a detailed, introductory exposition of the Metropolis-Hastings
	algorithm, a powerful Markov chain method to simulate multivariate
	distributions. A simple, intuitive derivation of this method is given
	along with guidance on implementation. Also discussed are two applications
	of the algorithm, one for implementing acceptance-rejection sampling
	when a blanketing function is not available and the other for implementing
	the algorithm with block-at-a-time scans. In the latter situation,
	many different algorithms, including the Gibbs sampler, are shown
	to be special cases of the Metropolis-Hastings algorithm. The methods
	are illustrated with examples.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Casella1992,
  author = {Casella, G. and George, E. I.},
  title = {Explaining the Gibbs sampler},
  journal = {American Statistician},
  year = {1992},
  volume = {46},
  pages = {167-174},
  number = {3},
  abstract = {Computer-intensive algorithms, such as the Gibbs sampler, have become
	increasingly popular statistical tools, both in applied and theoretical
	work. The properties of such algorithms, however, may sometimes not
	be obvious. Here we give a simple explanation of how and why the
	Gibbs sampler works. We analytically establish its properties in
	a simple case and provide insight for more complicated cases. There
	are also a number of examples.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Sullivan1998,
  author = {Sullivan, P. F. and Kessler, R. C. and Kendler, K. S.},
  title = {Latent class analysis of lifetime depressive symptoms in the National
	Comorbidity Survey},
  journal = {American Journal of Psychiatry},
  year = {1998},
  volume = {155},
  pages = {1398-1406},
  number = {10},
  abstract = {Objective: Although clinical trials have documented the importance
	of identifying individuals with major depression with atypical features,
	there are fewer epidemiological data; In a prior report, the authors
	used latent class analysis (LCA) to identify a distinctive atypical
	depressive subtype; they sought to replicate that finding in the
	current study. Method: Using the National Comorbidity Survey data,
	the authors applied LCA to 14 DSM-III-R major depressive symptoms
	in the participants' lifetime worst episodes (N = 2,836), Validators
	of class membership included depressive disorder characteristics,
	syndrome consequences, demography, comorbidity, personality/attitudes,
	and parental psychiatric history. Results: The best-Silting LCA solution
	had six classes. Four were combinations of atypicality and severity:
	severe atypical, mild atypical, severe typical, and mild typical.
	Syndrome severity (severe atypical and typical versus mild atypical
	and typical classes) was associated with a pronounced pattern of
	more and longer episodes, worse syndrome consequences, increased
	psychiatric comorbidity, more deviant personality and attitudes,
	and parental alcohol/drug use disorder. Syndrome atypicality (severe
	and mild atypical versus severe and mild typical classes) was associated
	with decreased syndrome consequences, comorbid conduct disorder and
	social phobia, higher interpersonal dependency and lower self-esteem,
	and parental alcohol/drug use disorder. Conclusions: As in prior
	reports,the atypical subtype of depression can be identified in epidemiological
	samples and, like typical depression, exists in mild and severe variants.
	Atypical depressive subtypes were characterized by several distinctive
	features. However, the correspondence between epidemiologically derived
	typologies of atypical depression and DSM-IV major depression with
	atypical features is not yet known.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Richardson1997,
  author = {Richardson, S. and Green, P. J.},
  title = {On Bayesian analysis of mixtures with an unknown number of components},
  journal = {Journal of the Royal Statistical Society Series B-Methodological},
  year = {1997},
  volume = {59},
  pages = {731-758},
  number = {4},
  abstract = {New methodology for fully Bayesian mixture analysis is developed,
	making use of reversible jump Markov chain Monte Carlo methods that
	are capable of jumping between the parameter subspaces corresponding
	to different numbers of components in the mixture. A sample from
	the full joint distribution of all unknown variables is thereby generated,
	and this can be used as a basis for a thorough presentation of many
	aspects of the posterior distribution. The methodology is applied
	here to the analysis of univariate normal mixtures, using a hierarchical
	prior model that offers an approach to dealing with weak prior information
	while avoiding the mathematical pitfalls of using improper priors
	in the mixture context.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Fraley1999,
  author = {Fraley, C. and Raftery, A. E.},
  title = {MCLUST: Software for model-based cluster analysis},
  journal = {Journal of Classification},
  year = {1999},
  volume = {16},
  pages = {297-306},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{DeBacker1999,
  author = {De Backer, S. and Scheunders, P.},
  title = {A competitive elliptical clustering algorithm},
  journal = {Pattern Recognition Letters},
  year = {1999},
  volume = {20},
  pages = {1141-1147},
  number = {11-13},
  abstract = {This paper introduces a new learning algorithm for on-line ellipsoidal
	clustering. The algorithm is based on the competitive clustering
	scheme extended by two specific features. Elliptical clustering is
	accomplished by efficiently incorporating the Mahalanobis distance
	measure into the learning rules, and underutilization of smaller
	dusters is avoided by incorporating a frequency-sensitive term. Experiments
	are conducted to demonstrate the usefulness of the algorithm on artificial
	data-sets as well as on the problem of texture segmentation. (C)
	1999 Elsevier Science B.V. All rights reserved.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Fraley2002,
  author = {Fraley, C. and Raftery, A. E.},
  title = {Model-based clustering, discriminant analysis, and density estimation},
  journal = {Journal of the American Statistical Association},
  year = {2002},
  volume = {97},
  pages = {611-631},
  number = {458},
  abstract = {Cluster analysis is the automated search for groups of related observations
	in a dataset. Most clustering done in practice is based largely on
	heuristic but intuitively reasonable procedures, and most clustering
	methods available in commercial software are also of this type. However,
	there is little systematic guidance associated with these methods
	for solving important practical questions that arise in cluster analysis,
	such as how many clusters are there, which clustering method should
	be used, and how should outliers be handled. We review a general
	methodology for model-based clustering that provides a principled
	statistical approach to these issues. We also show that this can
	be useful for other problems in multivariate analysis, such as discriminant
	analysis and multivariate density estimation. We give examples from
	medical diagnosis, minefield detection, cluster recovery from noisy
	data, and spatial density estimation. Finally, we mention limitations
	of the methodology and discuss recent developments in model-based
	clustering for non-Gaussian data, high-dimensional datasets, large
	datasets, and Bayesian estimation.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@TECHREPORT{Fraley2006,
  author = {Fraley, C. F. and Raftery, A. E.},
  title = {Technical Report 504. MCLUST Version 3 for R: Normal mixture modeling
	and model-based clustering},
  institution = {Department of Statistics, University of Washington},
  year = {2006},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Dempster1977,
  author = {Dempster, A. P. and Laird, N. M. and Rubin, D. B.},
  title = {Maximum Likelihood from Incomplete Data via the EM Algorithm},
  journal = {Journal of the Royal Statistical Society. Series B (Methodological)},
  year = {1977},
  volume = {39},
  pages = {1-38},
  number = {1},
  abstract = {A broadly applicable algorithm for computing maximum likelihood estimates
	from incomplete data is presented at various levels of generality.
	Theory showing the monotone behaviour of the likelihood and convergence
	of the algorithm is derived. Many examples are sketched, including
	missing value situations, applications to grouped, censored or truncated
	data, finite mixture models, variance component estimation, hyperparameter
	estimation, iteratively reweighted least squares and factor analysis.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Ward1963,
  author = {Ward, Joe H., Jr.},
  title = {Hierarchical Grouping to Optimize an Objective Function},
  journal = {Journal of the American Statistical Association},
  year = {1963},
  volume = {58},
  pages = {236-244},
  number = {301},
  abstract = {A procedure for forming hierarchical groups of mutually exclusive
	subsets, each of which has members that are maximally similar with
	respect to specified characteristics, is suggested for use in large-scale
	(<latex>$n > 100$</latex>) studies when a precise optimal solution
	for a specified number of groups is not practical. Given n sets,
	this procedure permits their reduction to n - 1 mutually exclusive
	sets by considering the union of all possible n(n - 1)/2 pairs and
	selecting a union having a maximal value for the functional relation,
	or objective function, that reflects the criterion chosen by the
	investigator. By repeating this process until only one group remains,
	the complete hierarchical structure and a quantitative estimate of
	the loss associated with each stage in the grouping can be obtained.
	A general flowchart helpful in computer programming and a numerical
	example are included.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Fowlkes1983,
  author = {Fowlkes, E. B. and Mallows, C. L.},
  title = {A method for comparing 2 hierarchical clusterings},
  journal = {Journal of the American Statistical Association},
  year = {1983},
  volume = {78},
  pages = {553-569},
  number = {383},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Morey1984,
  author = {Morey, L. C. and Agresti, A.},
  title = {The measurement of classification agreement - an adjustment to the
	Rand statistic for chance agreement},
  journal = {Educational and Psychological Measurement},
  year = {1984},
  volume = {44},
  pages = {33-37},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Klastorin1985,
  author = {Klastorin, T. D.},
  title = {The para-median problem for cluster-analysis - a comparative test
	using the mixture model approach},
  journal = {Management Science},
  year = {1985},
  volume = {31},
  pages = {84-95},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Hubert1985,
  author = {Hubert, L. and Arabie, P.},
  title = {Comparing partitions},
  journal = {Journal of Classification},
  year = {1985},
  volume = {2},
  pages = {193-218},
  number = {2-3},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Fisher1958,
  author = {Fisher, W. D.},
  title = {On grouping for maximum homogeneity},
  journal = {Journal of the American Statistical Association},
  year = {1958},
  volume = {53},
  pages = {789-798},
  number = {284},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Ball1967,
  author = {Ball, G. H. and Hall, D. J.},
  title = {A clustering technique for summarizing multivariate data},
  journal = {Behavioral Science},
  year = {1967},
  volume = {12},
  pages = {153-\&},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Mantel1967,
  author = {Mantel, N.},
  title = {Detection of disease clustering and a generalized regression approach},
  journal = {Cancer Research},
  year = {1967},
  volume = {27},
  pages = {209-\&},
  number = {2P1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Rubin1967,
  author = {Rubin, J.},
  title = {Optimal classification into groups - an approach for solving taxonomy
	problem},
  journal = {Journal of Theoretical Biology},
  year = {1967},
  volume = {15},
  pages = {103-\&},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Rand1971,
  author = {Rand, W. M.},
  title = {Objective criteria for evaluation of clustering methods},
  journal = {Journal of the American Statistical Association},
  year = {1971},
  volume = {66},
  pages = {846-\&},
  number = {336},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Brennan1974,
  author = {Brennan, R. L. and Light, R. J.},
  title = {Measuring agreement when 2 observers classify people into categories},
  journal = {British Journal of Mathematical \& Statistical Psychology},
  year = {1974},
  volume = {27},
  pages = {154-163},
  number = {NOV},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.25}
}

@ARTICLE{Chandra1977,
  author = {Chandra, S.},
  title = {MIXTURES OF PROBABILITY DISTRIBUTIONS},
  journal = {Scandinavian Journal of Statistics},
  year = {1977},
  volume = {4},
  pages = {105-112},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@ARTICLE{Kasahara2009,
  author = {Kasahara, H. and Shimotsu, K.},
  title = {Nonparametric Identification of Finite Mixture Models of Dynamic
	Discrete Choices},
  journal = {Econometrica},
  year = {2009},
  volume = {77},
  pages = {135-175},
  number = {1},
  abstract = {In dynamic discrete choice analysis, controlling for unobserved heterogeneity
	is an important issue, and finite mixture models provide flexible
	ways to account for it. This paper studies nonparametric identifiability
	of type probabilities and type-specific component distributions in
	finite mixture models of dynamic discrete choices. We derive sufficient
	conditions for nonparametric identification for various finite mixture
	models of dynamic discrete choices used in applied work under different
	assumptions on the Markov property, stationarity, and type-invariance
	in the transition process. Three elements emerge as the important
	determinants of identification: the time-dimension of panel data,
	the number of values the covariates can take, and the heterogeneity
	of the response of different types to changes in the covariates.
	For example, in a simple case where the transition function is type-invariant,
	a time-dimension of T = 3 is sufficient for identification, provided
	that the number of values the covariates can take is no smaller than
	the number of types and that the changes in the covariates induce
	sufficiently heterogeneous variations in the choice probabilities
	across types. Identification is achieved even when state dependence
	is present if a model is stationary first-order Markovian and the
	panel has a moderate time-dimension (T >= 6).},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@ARTICLE{Yakowitz1968,
  author = {Yakowitz, S. J. and Spragins, J. D.},
  title = {On identifiability of finite mixtures},
  journal = {Annals of Mathematical Statistics},
  year = {1968},
  volume = {39},
  pages = {209-\&},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@ARTICLE{Wolfe1970,
  author = {Wolfe, J. H.},
  title = {Pattern clustering by multivariate mixture analysis},
  journal = {Multivariate Behavioral Research},
  year = {1970},
  volume = {5},
  pages = {329-350},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@ARTICLE{Teicher1963,
  author = {Teicher, H.},
  title = {Identifiability of finite mixtures},
  journal = {Annals of Mathematical Statistics},
  year = {1963},
  volume = {34},
  pages = {1265-\&},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@ARTICLE{Teicher1961,
  author = {Teicher, H.},
  title = {Identifiability of mixtures},
  journal = {Annals of Mathematical Statistics},
  year = {1961},
  volume = {32},
  pages = {244-248},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@ARTICLE{Stefankovic2007,
  author = {Stefankovic, D. and Vigoda, E.},
  title = {Phylogeny of mixture models: Robustness of maximum likelihood and
	non-identifiable distributions},
  journal = {Journal of Computational Biology},
  year = {2007},
  volume = {14},
  pages = {156-189},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@ARTICLE{Morris2007,
  author = {Morris, J. N.},
  title = {Uses of epidemiology (vol 2, pg 395-401, 1955)},
  journal = {International Journal of Epidemiology},
  year = {2007},
  volume = {36},
  pages = {1165-1172},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@BOOK{Morris1975,
  title = {Uses of epidemiology},
  publisher = {Churchill Livingstone : distributed in the U.S.A. by Longman},
  year = {1975},
  author = {Morris, Jeremy Noah},
  address = {Edinburgh ; New York},
  edition = {3d},
  keywords = {Epidemiology. Epidemiology.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@ARTICLE{Morris1955,
  author = {Morris, J. N.},
  title = {Uses of epidemiology},
  journal = {British Medical Journal},
  year = {1955},
  volume = {2},
  pages = {395-401},
  number = {AUG13},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@ELECTRONIC{CramerRule,
  author = {Weisstein, E. W.},
  title = {Cramer's Rule.},
  organization = {MathWorld--A Wolfram Web Resource},
  url = {http://mathworld.wolfram.com/CramersRule.html},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@ELECTRONIC{singular,
  author = {Weisstein, E. W.},
  title = {Singular Matrix.},
  organization = {MathWorld--A Wolfram Web Resource},
  url = {http://mathworld.wolfram.com/SingularMatrix.html},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@INPROCEEDINGS{Park2004,
  author = {Park, J. W. and Logue, M. and Ni, J. and Cremer, J. and Segre, A.
	and Vieland, V.},
  title = {Scientific visualization of multidimensional data: Genetic likelihood
	visualization},
  booktitle = {International Conference on High Performance Computing and Its Applications},
  year = {2004},
  editor = {Zhang, W. Chen Z. Glowinski R. Tong W.},
  pages = {403-408},
  address = {Shanghai, PEOPLES R CHINA},
  abstract = {Although many computer graphic technologies have been developed for
	visualizing multidimensional multivariate data, the scientific visualization
	used by research scientists to interpret genetics data is very promising
	technique. In this paper, we present our research in a scientific
	visualization on linkage analysis data to enhance the performance
	or the efficiency of genetic likelihood research.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@BOOK{Eliason1993,
  title = {Maximum likelihood estimation : logic and practice},
  publisher = {Sage},
  year = {1993},
  author = {Eliason, Scott R.},
  series = {Sage university papers series. Quantitative applications in the social
	sciences},
  address = {Newbury Park, Calif.},
  keywords = {Social sciences Statistical methods. Estimation theory.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@ARTICLE{Groeneveld1990,
  author = {Groeneveld, E. and Kovac, M.},
  title = {A note on multiple solutions in multivariate restricted maximum-likelihood
	covariance component estimation},
  journal = {Journal of Dairy Science},
  year = {1990},
  volume = {73},
  pages = {2221-2229},
  number = {8},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@INPROCEEDINGS{Atwood1992,
  author = {Atwood, L. D. and Kammerer, C. M. and Mitchell, B. D.},
  title = {Exploring the HDL likelihood surface},
  booktitle = {Genetic Analysis Workshop 8: Issues in the Analysis of Complex Diseases
	and Their Risk Factors ( GAW8 )},
  year = {1992},
  pages = {641-645},
  address = {Watsonville, Ca},
  abstract = {Using random initial parameter estimates, three segregation analysis
	models of the inheritance of HDL2 in the Berkeley GAW8 data set were
	maximized 5000 times each. Initial parameter estimates were assumed
	to be uniformly distributed on intervals formed by parameter boundaries.
	The three models were unrestricted, environmental, and Mendelian
	regressive type A models. Likelihood ratio tests of the global maxima
	rejected the Mendelian model and accepted the environmental model.
	However, tests using local maxima accepted the Mendelian model and
	both rejected and accepted the environmental model. Patterns among
	the initial parameter estimates of convergent runs were examined
	to develop empirical rules to increase the frequency of convergence.
	These rules were tested using data on apoAI in the Berkeley GAW8
	data set. (C) 1993 Wiley-Liss, Inc.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.26}
}

@BOOK{McCutcheon1987,
  title = {Latent class analysis},
  publisher = {Sage Publications},
  year = {1987},
  author = {McCutcheon, Allan L.},
  series = {Sage University papers series. Quantitative applications in the social
	sciences},
  address = {Newbury Park},
  keywords = {Latent structure analysis. Latent variables.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.27}
}

@BOOK{Hagenaars2002,
  title = {Applied latent class analysis},
  publisher = {Cambridge University Press},
  year = {2002},
  author = {Hagenaars, Jacques A. and McCutcheon, Allan L.},
  address = {Cambridge ; New York},
  keywords = {Latent structure analysis. Latent variables.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.27}
}

@BOOK{Gordon1999,
  title = {Classification},
  publisher = {Chapman \& Hall/CRC},
  year = {1999},
  author = {Gordon, A. D.},
  series = {Monographs on statistics and applied probability},
  address = {Boca Raton},
  edition = {2nd},
  keywords = {Discriminant analysis. Cluster analysis.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.28}
}

@BOOK{Gordon1981,
  title = {Classification : methods for the exploratory analysis of multivariate
	data},
  publisher = {Chapman and Hall},
  year = {1981},
  author = {Gordon, A. D.},
  series = {Monographs on applied probability and statistics},
  address = {London ; New York},
  keywords = {Discriminant analysis. Cluster analysis.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.28}
}

@INBOOK{Bock1996,
  chapter = {Probability models and hypothesis testing in partitioning cluster
	analysis},
  pages = {377-453},
  title = {Clustering and Classification},
  publisher = {World Scientific Publ.},
  year = {1996},
  editor = {Arabie P., Hubert L. J., De Soete G.},
  author = {Bock, H. H.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.28}
}

@ARTICLE{Espeland1989,
  author = {Espeland, M. A. and Handelman, S. L.},
  title = {Using latent class models to characterize and assess relative error
	in discrete measurements},
  journal = {Biometrics},
  year = {1989},
  volume = {45},
  pages = {587-599},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Ganesalingam1989,
  author = {Ganesalingam, S.},
  title = {Classification and mixture approaches to clustering via maximum-likelihood},
  journal = {Applied Statistics-Journal of the Royal Statistical Society Series
	C},
  year = {1989},
  volume = {38},
  pages = {455-466},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Trauwaert1991,
  author = {Trauwaert, E. and Kaufman, L. and Rousseeuw, P.},
  title = {Fuzzy clustering algorithms based on the maximum-likelihood principle},
  journal = {Fuzzy Sets and Systems},
  year = {1991},
  volume = {42},
  pages = {213-227},
  number = {2},
  abstract = {A number of hard clustering algorithms have been shown to be derivable
	from the maximum likelihood principle. The only corresponding fuzzy
	algorithm are the well known fuzzy k-means or FUZZY ISODATA of Dunn
	and its generalizations by Bezdek and by Gustafson and Kessel. The
	authors show how to generate two other fuzzy algorithms which are
	the analogous of known hard algorithms: the minimization of the fuzzy
	determinant and of the product of fuzzy determinants. By comparison
	between the hard and fuzzy methods it appears that the latter yield
	more often the global optimum, rather than merely a local optimum.
	This result and the comparison between the different algorithms,
	together with their specific domains of application, are illustrated
	by a few numerical examples.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Celeux1991,
  author = {Celeux, G. and Govaert, G.},
  title = {Clustering criteria for discrete-data and latent class models},
  journal = {Journal of Classification},
  year = {1991},
  volume = {8},
  pages = {157-176},
  number = {2},
  abstract = {We show that a well-known clustering criterion for discrete data,
	the information criterion, is closely related to the classification
	maximum likelihood criterion for the latent class model. This relation
	can be derived from the Bryant-Windham construction. Emphasis is
	placed on binary clustering criteria which are analyzed under the
	maximum likelihood approach for different multivariate Bernoulli
	mixtures. This alternative form of criterion reveals non-apparent
	aspects of clustering techniques. All the criteria discussed can
	be optimized with the alternating optimization algorithm. Some illustrative
	applications are included.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Arabie1992,
  author = {Arabie, P. and Hubert, L. J.},
  title = {Combinatorial data analysis},
  journal = {Annual Review of Psychology},
  year = {1992},
  volume = {43},
  pages = {169-203},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Celeux1992,
  author = {Celeux, G. and Govaert, G.},
  title = {A classification EM algorithm for clustering and 2 stochastic versions},
  journal = {Computational Statistics \& Data Analysis},
  year = {1992},
  volume = {14},
  pages = {315-332},
  number = {3},
  abstract = {Setting the optimization-based clustering methods under the classification
	maximum likelihood approach, we define and study a general Classification
	EM algorithm. Then, we derive from this algorithm two stochastic
	algorithms, incorporating random perturbations, to reduce the initial-position
	dependence of the classical optimization clustering algorithms. Numerical
	experiments, reported for the variance criterion, show that both
	stochastic algorithms perform well compared with the standard k-means
	algorithm which is a particular version of the Classification EM
	algorithm.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Yang1993,
  author = {Yang, M. S.},
  title = {On a class of fuzzy classification maximum-likelihood procedures},
  journal = {Fuzzy Sets and Systems},
  year = {1993},
  volume = {57},
  pages = {365-375},
  number = {3},
  abstract = {Classification Maximum Likelihood (CML) procedure is a remarkable
	mixture of maximum likelihood approach to clustering. This has been
	well documented in the book of McLachlan and Basford, In this paper,
	we make the fuzzy extension of the CML procedure. Based on this class
	of fuzzy CML procedures, we extend the fuzzy clustering algorithms
	of Trauwaert, Kaufman and Rousseeuw by adding a penalty term. Especially,
	we derive a generalized type of fuzzy c-means (FCM) clustering algorithms,
	called the penalized FCM clustering algorithms. Then we create some
	asymptotic behaviors of these penalized FCM procedures. By doing
	some numerical examples we find that the penalized FCM is more meaningful
	and effective than FCM.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Qu1996,
  author = {Qu, Y. S. and Tan, M. and Kutner, M. H.},
  title = {Random effects models in latent class analysis for evaluating accuracy
	of diagnostic tests},
  journal = {Biometrics},
  year = {1996},
  volume = {52},
  pages = {797-810},
  number = {3},
  abstract = {When the results of a reference (or gold standard) test are missing
	or not error-free, the accuracy of diagnostic tests is often assessed
	through latent class models with two latent classes, representing
	diseased or nondiseased status. Such models, however, require that
	conditional on the true disease status, the tests are statistically
	independent, an assumption often violated in practice. Consequently,
	the model generally fits the data poorly. In this paper, we develop
	a general latent class model with random effects to model the conditional
	dependence among multiple diagnostic tests (or readers). We also
	develop a graphical method for checking whether or not the conditional
	dependence is of concern and for identifying the pattern of the correlation.
	Using the random-effects model and the graphical method, a simple
	adequate model that is easy to interpret can be obtained. The methods
	are illustrated with three examples from the biometric literature.
	The proposed methodology is also applicable when the true disease
	status is indeed known and conditional dependence could well be present.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Celeux1996,
  author = {Celeux, G. and Soromenho, G.},
  title = {An entropy criterion for assessing the number of clusters in a mixture
	model},
  journal = {Journal of Classification},
  year = {1996},
  volume = {13},
  pages = {195-212},
  number = {2},
  abstract = {In this paper, we consider an entropy criterion to estimate the number
	of clusters arising from a mixture model. This criterion is derived
	from a relation linking the likelihood and the classification likelihood
	of a mixture. Its performance is investigated through Monte Carlo
	experiments, and it shows favorable results compared to other classical
	criteria.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Yang2005,
  author = {Yang, M. S. and Yu, N. Y.},
  title = {Estimation of parameters in latent class models using fuzzy clustering
	algorithms},
  journal = {European Journal of Operational Research},
  year = {2005},
  volume = {160},
  pages = {515-531},
  number = {2},
  abstract = {A mixture approach to clustering is an important technique in cluster
	analysis. A mixture of multivariate multinormal distributions is
	usually used to analyze categorical data with latent class model.
	The parameter estimation is an important step for a mixture distribution.
	Described here are four approaches to estimating the parameters of
	a mixture of multivariate multinomial distributions. The first approach
	is an extended maximum likelihood (ML) method. The second approach
	is based on the well-known expectation maximization (EM) algorithm.
	The third approach is the classification maximum likelihood (CML)
	algorithm. In this paper, we propose a new approach using the so-called
	fuzzy class model and then create the fuzzy classification maximum
	likelihood (FCML) approach for categorical data. The accuracy, robustness
	and effectiveness of these four types of algorithms for estimating
	the parameters of multivariate binomial mixtures are compared using
	real empirical data and samples drawn from the multivariate binomial
	mixtures of two classes. The results show that the proposed FCML
	algorithm presents better accuracy, robustness and effectiveness.
	Overall, the FCML algorithm has the superiority over the ML, EM and
	CML algorithms. Thus, we recommend FCML as another good tool for
	estimating the parameters of mixture multivariate multinomial models.
	(C) 2003 Elsevier B.V. All rights reserved.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Bock1996a,
  author = {Bock, H. H.},
  title = {Probabilistic models in cluster analysis},
  journal = {Computational Statistics \& Data Analysis},
  year = {1996},
  volume = {23},
  pages = {5-28},
  number = {1},
  abstract = {This paper discusses cluster analysis in a probabilistic and inferential
	framework as opposed to more exploratory, heuristic or algorithmic
	approaches, It presents a broad survey on probabilistic models for
	partition-type, hierarchical and tree-like clustering structures
	and points to the relevant literature. It is shown how suitable clustering
	criteria or grouping methods may be derived from these models in
	the case of vector-valued data, dissimilarity matrices and similarity
	relations. In particular, we discuss hypothesis testing for homogeneity
	or for a grouping structure, the asymptotic distribution of test
	statistics, the use of random graph theory and combinatorial methods
	for simulating random dendrograms. Our presentation of hierarchies
	includes, e.g., Markovian branching processes and phylogenetic inference
	based on molecular sequence data.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Huang2004,
  author = {Huang, G. H. and Bandeen-Roche, K.},
  title = {Building an identifiable latent class model with covariate effects
	on underlying and measured variables},
  journal = {Psychometrika},
  year = {2004},
  volume = {69},
  pages = {5-32},
  number = {1},
  abstract = {In recent years, latent class models have proven useful for analyzing
	relationships between measured multiple indicators and covariates
	of interest. Such models summarize shared features of the multiple
	indicators as an underlying categorical variable, and the indicators'
	substantive associations with predictors are built directly and indirectly
	in unique model parameters. In this paper, we provide a detailed
	study on the theory and application of building models that allow
	mediated relationships between primary predictors and latent class
	membership, but that also allow direct effects of secondary covariates
	on the indicators themselves. Theory for model identification is
	developed. We detail an Expectation-Maximization algorithm for parameter
	estimation, standard error calculation, and convergent properties.
	Comparison of the proposed model with models underlying existing
	latent class modeling software is provided. A detailed analysis of
	how visual impairments affect older persons' functioning requiring
	distance vision is used for illustration.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Reboussin2006,
  author = {Reboussin, B. A. and Song, E. Y. and Shrestha, A. and Lohman, K.
	K. and Wolfson, M.},
  title = {A latent class analysis of underage problem drinking: Evidence from
	a community sample of 16-20 year olds},
  journal = {Drug and Alcohol Dependence},
  year = {2006},
  volume = {83},
  pages = {199-209},
  number = {3},
  abstract = {The aim of this paper is to shed light on the nature, of underage
	problem drinking by using an empirically based method to characterize
	the variation in patterns of drinking in a community sample of underage
	drinkers. A total of 4056 16-20-year-old current drinkers from 212
	communities in the US were surveyed by telephone as part of the National
	Evaluation of the Enforcing Underage Drinking Laws (EUDL) Program.
	Latent class models were used to create homogenous groups of drinkers
	with similar drinking patterns defined by multiple indicators of
	drinking behaviors and alcohol-related problems. Two types of underage
	problem drinkers were identified; risky drinkers (30%) and regular
	drinkers (27%). The most prominent behaviors among both types of
	underage problem drinkers were binge drinking and getting drunk.
	Being male, other drug use, early onset drinking and beliefs about
	friends drinking and getting drunk were all associated with an increased
	risk of being a problem drinker after adjustment for other factors.
	Beliefs that most friends drink and current marijuana use were the
	strongest predictors of both risky problem drinking (OR = 4.0; 95%
	CI = 3.1, 5.1 and OR = 4.0; 95% CI = 2.8, 5.6, respectively) and
	regular problem drinking (OR = 10.8; 95% CI 7.0, 16.7 and OR = 10.2;
	95% CI = 6.9, 15.2). Young adulthood (ages 18-20) was significantly
	associated with regular problem drinking but not risky problem drinking.
	The belief that most friends get drunk weekly was the strongest discriminator
	of risky and regular problem drinking patterns (OR = 5.3; 95% CI
	= 3.9, 7.1). These findings suggest that underage problem drinking
	is most strongly characterized by heavy drinking behaviors which
	can emerge in late adolescence and underscores its association with
	perceptions regarding friends drinking behaviors and illicit drug
	use. (c) 2005 Elsevier Ireland Ltd. All rights reserved.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Reboussin2006b,
  author = {Reboussin, B. A. and Anthony, J. C.},
  title = {Is there epidemiological evidence to support the idea that a cocaine
	dependence syndrome emerges soon after onset of cocaine use?},
  journal = {Neuropsychopharmacology},
  year = {2006},
  volume = {31},
  pages = {2055-2064},
  number = {9},
  abstract = {The present study uses latent class methods and multiple regression
	to shed light on hypothesized cocaine dependence syndromes experienced
	by community residents, who initiated cocaine use within 24 months
	of survey assessment, and explores possible variation in risk. Identified
	within public use data files from the United States National Household
	Surveys on Drug Abuse (NHSDA), and with assessments completed between
	1995 and 1998, the study sample consists of 927 recent-onset cocaine
	users, defined as having initiated cocaine use no more than 24 months
	prior to assessment (approximate median elapsed time since onset
	of use similar to 12-13 months). The NHSDA included items to assess
	seven clinical features often associated with cocaine dependence,
	which were used in latent class modeling. Empirically derived latent
	classes, in conjunction with prior theory, tend to support a three-class
	solution, according to which 4% of recent-onset users are members
	of a class that resembles the DSM-IV cocaine dependence syndrome
	(mean: 5.4 clinical features (CF)); 16% might be in a cocaine dependence
	prodrome (mean: 2.4 CF); 80% of recent-onset cocaine users had few
	or no clinical features (mean < 1 CF). Results from latent class
	regressions indicate that susceptibility to rapid transition from
	first cocaine use to onset of the LCA-assigned cocaine dependence
	syndrome might depend upon whether the user starts smoking crack-cocaine
	and, independently, age at first cocaine use.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Reboussin2006a,
  author = {Reboussin, B. A. and Lohman, K. K. and Wolfson, M.},
  title = {Modeling adolescent drug use patterns in cluster-unit trials with
	multiple sources of correlation using robust latent class regressions},
  journal = {Annals of Epidemiology},
  year = {2006},
  volume = {16},
  pages = {850-859},
  number = {11},
  abstract = {PURPOSE: The purpose of the study is to examine variation in adolescent
	drug-use patterns by using latent class regression analysis and evaluate
	the properties of an estimating-equations approach under different
	cluster-unit trial designs. METHODS: A set of second-order estimating
	equations for latent class models under the cluster-unit trial design
	are proposed. This approach models the correlation within subclusters
	(drug-use behaviors), but ignores the correlation within clusters
	(communities). A robust covariance estimator is proposed that accounts
	for within-cluster correlation. Performance of this approach is addressed
	through a Monte Carlo simulation study, and practical implications
	are illustrated by using data from the National Evaluation of the
	Enforcing Underage Drinking Laws Randomized Community Trial. RESULTS:
	The example shows that the proposed method provides useful information
	about the heterogeneous nature of drug use by identifying two subtypes
	of adolescent problem drinkers. A Monte Carlo simulation study supports
	the proposed estimation method by suggesting that the latent class
	model parameters were unbiased for 30 or more clusters. Consistent
	with other studies of generalized estimating equation (GEE) estimators,
	the robust covariance estimator tended to underestimate the true
	variance of regression parameters, but the degree of inflation in
	the test size was relatively small for 70 clusters and only slightly
	inflated for 30 clusters. CONCLUSIONS: The proposed model for studying
	adolescent drug use provides an alternative to standard diagnostic
	criteria, focusing on the nature of the drug-use profile, rather
	than relying on univariate symptom counts. The second-order GEE-type
	estimation procedure provided a computationally feasible approach
	that performed well for a moderate number of clusters and was consistent
	with prior studies of GEE under the generalized linear model framework.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Harper1972,
  author = {Harper, D.},
  title = {Local dependence latent structure models},
  journal = {Psychometrika},
  year = {1972},
  volume = {37},
  pages = {53-\&},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Ganesalingam1980,
  author = {Ganesalingam, S. and McLachlan, G. J.},
  title = {A comparison of the mixture and classification approaches to cluster
	analysis},
  journal = {Communications in Statistics Part a-Theory and Methods},
  year = {1980},
  volume = {9},
  pages = {923-933},
  number = {9},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Suppes1981,
  author = {Suppes, P. and Zanotti, M.},
  title = {When are probabilistic explanations possible},
  journal = {Synthese},
  year = {1981},
  volume = {48},
  pages = {191-199},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Windham1985,
  author = {Windham, M. P.},
  title = {Numerical classification of proximity data with assignment measures},
  journal = {Journal of Classification},
  year = {1985},
  volume = {2},
  pages = {157-172},
  number = {2-3},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Basford1985,
  author = {Basford, K. E. and McLachlan, G. J.},
  title = {Likelihood estimation with normal mixture models},
  journal = {Applied Statistics-Journal of the Royal Statistical Society Series
	C},
  year = {1985},
  volume = {34},
  pages = {282-289},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Quinn1987,
  author = {Quinn, B. G. and McLachlan, G. J. and Hjort, N. L.},
  title = {A note on the Aitkin-Rubin approach to hypothesis-testing in mixture-models},
  journal = {Journal of the Royal Statistical Society Series B-Methodological},
  year = {1987},
  volume = {49},
  pages = {311-314},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Hagenaars1988,
  author = {Hagenaars, J. A.},
  title = {Latent structure models with direct effects between indicators -
	local dependence models},
  journal = {Sociological Methods \& Research},
  year = {1988},
  volume = {16},
  pages = {379-405},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Bensmail1996,
  author = {Bensmail, H. and Celeux, G.},
  title = {Regularized Gaussian discriminant analysis through eigenvalue decomposition},
  journal = {Journal of the American Statistical Association},
  year = {1996},
  volume = {91},
  pages = {1743-1748},
  number = {436},
  abstract = {Friedman proposed a regularization technique (RDA) of discriminant
	analysis in the Gaussian framework. RDA uses two regularization parameters
	to design an intermediate classifier between the linear, the quadratic,
	and the nearest-means classifiers. In this article we propose an
	alternative approach, called EDDA, that is based on the reparameterization
	of the covariance matrix [Sigma(k)] of a group G(k) in terms of its
	eigenvalue decomposition Sigma(k) = lambda(k)D(k)A(k)D(k)', where
	lambda(k) specifies the volume of density contours of G(k), the diagonal
	matrix of eigenvalues specifies its shape, and the eigenvectors specify
	its orientation. Variations on constraints concerning volumes, shapes,
	and orientations lambda(k), A(k), and D-k lead to 14 discrimination
	models of interest. For each model, we derived the normal theory
	maximum likelihood parameter estimates. Our approach consists of
	selecting a model by minimizing the sample-based estimate of future
	misclassification risk by cross-validation. Numerical experiments
	on simulated and real data show favorable behavior of this approach
	compared to RDA.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Bryant1988,
  author = {Bryant, P.},
  title = {On characterizing optimization-based clustering methods},
  journal = {Journal of Classification},
  year = {1988},
  volume = {5},
  pages = {81-84},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Bensmail1997,
  author = {Bensmail, H. and Celeux, G. and Raftery, A. E. and Robert, C. P.},
  title = {Inference in model-based cluster analysis},
  journal = {Statistics and Computing},
  year = {1997},
  volume = {7},
  pages = {1-10},
  number = {1},
  abstract = {A new approach to cluster analysis has been introduced based on parsimonious
	geometric modelling of the within-group covariance matrices in a
	mixture of multivariate normal distributions, using hierarchical
	agglomeration and iterative relocation. It works well and is widely
	used via the MCLUST software available in S-PLUS and StatLib. However,
	it has several limitations: there is no assessment of the uncertainty
	about the classification, the partition can be suboptimal, parameter
	estimates are biased, the shape matrix has to be specified by the
	user, prior group probabilities are assumed to be equal, the method
	for choosing the number of groups is based on a crude approximation,
	and no formal way of choosing between the various possible models
	is included. Here, we propose a new approach which overcomes all
	these difficulties. It consists of exact Bayesian inference via Gibbs
	sampling, and the calculation of Bayes factors (for choosing the
	model and the number of groups) from the output using the Laplace-Metropolis
	estimator. It works well in several real and simulated examples.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Lin1997,
  author = {Lin, T. H. and Dayton, C. M.},
  title = {Model selection information criteria for non-nested latent class
	models},
  journal = {Journal of Educational and Behavioral Statistics},
  year = {1997},
  volume = {22},
  pages = {249-264},
  number = {3},
  abstract = {Latent class models have been developed for assessment of hierarchic
	hierarchic relations in scaling and behavioral analysis. This article
	investigated the use of three model selection information criteria
	- Akaike AIC, Schwarz SIG, and Bozdogan CAIC - for non-nested models.
	In general, SIC and CAIC were superior to AIC for relatively simple
	models, whereas AIC was superior for more complex models, although
	accuracy was often quite low Sor such models. In addition, some effects
	were detected for error rates in the models.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Reboussin1999,
  author = {Reboussin, B. A. and Liang, K. Y. and Reboussin, D. M.},
  title = {Estimating equations for a latent transition model with multiple
	discrete indicators},
  journal = {Biometrics},
  year = {1999},
  volume = {55},
  pages = {839-845},
  number = {3},
  abstract = {This paper proposes a two-part model for studying transitions between
	health states over time when multiple, discrete health indicators
	are available. The includes a measurement model positing underlying
	latent health states and a transition model between latent health
	states over time. Full:maximum likelihood estimation procedures are
	computationally complex in this latent variable framework, making
	only a limited class of models feasible and estimation of standard
	errors problematic. For this reason, an estimating equations analogue
	of the pseudo-likelihood method for the parameters of interest, namely
	the transition model parameters, is considered. The finite sample
	properties of the proposed procedure are investigated through a simulation
	study and the importance of choosing strong indicators of the latent
	variable is demonstrated. The applicability of the methodology is
	illustrated with health survey data measuring disability in the elderly
	from the Longitudinal Study of Aging.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Roeder1999,
  author = {Roeder, K. and Lynch, K. G. and Nagin, D. S.},
  title = {Modeling uncertainty in latent class membership: A case study in
	criminology},
  journal = {Journal of the American Statistical Association},
  year = {1999},
  volume = {94},
  pages = {766-776},
  number = {447},
  abstract = {Social scientists are commonly interested in relating a latent trait
	(e.g., criminal tendency) to measurable individual covariates (e.g.,
	poor parenting) to understand what defines or perhaps causes the
	latent trait. In this article we develop an efficient and convenient
	method for answering such questions. The basic model presumes that
	two types of variables have been measured: response variables (possibly
	longitudinal) that partially determine the latent class membership,
	and covariates or risk factors that we wish to relate to these latent
	class variables. The model assumes that these observable variables
	are conditionally independent, given the latent class variable. We
	use a mixture model for the joint distribution of the observables.
	We apply this model to a longitudinal dataset assembled as part of
	the Cambridge Study of Delinquent Development to test a fundamental
	theory of criminal development. This theory holds that crime is committed
	by two distinct groups within the population: adolescent-limited
	offenders and life-course-persistent offenders. As these labels suggest,
	the two groups are distinguished by the longevity of their offending
	careers. The theory also predicts that life-course-persistent offenders
	are disproportionately comprised of individuals born with neurological
	deficits and reared by caregivers without the skills and resources
	to effectively socialize a difficult child.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Peel2000,
  author = {Peel, D. and McLachlan, G. J.},
  title = {Robust mixture modelling using the t distribution},
  journal = {Statistics and Computing},
  year = {2000},
  volume = {10},
  pages = {339-348},
  number = {4},
  abstract = {Normal mixture models are being increasingly used to model the distributions
	of a wide variety of random phenomena and to cluster sets of continuous
	multivariate data. However, for a set of data containing a group
	or groups of observations with longer than normal tails or atypical
	observations, the use of normal components may unduly affect the
	fit of the mixture model. In this paper, we consider a more robust
	approach by modelling the data by a mixture of t distributions. The
	use of the ECM algorithm to fit this t mixture model is described
	and examples of its use are given in the context of clustering multivariate
	data in the presence of atypical observations in the form of background
	noise.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Hennig2000,
  author = {Hennig, C.},
  title = {Identifiability of models for clusterwise linear regression},
  journal = {Journal of Classification},
  year = {2000},
  volume = {17},
  pages = {273-296},
  number = {2},
  abstract = {Identifiability of the parameters is a necessary condition for the
	existence of consistent estimators. In this paper the identifiability
	of the parameters of models for data generated by different linear
	regression distributions with Gaussian errors is investigated. It
	turns out that such models cause other identifiability problems than
	do simple Gaussian mixtures. This problem was heretofore ignored;
	thus there are no satisfying consistency proofs in this area. Three
	different models are treated: Finite mixture models with random and
	fixed covariates and a fixed partition model. Counterexamples and
	sufficient conditions for identifiability are given, including an
	example for nonidentifiable parameters with an invertible information
	matrix. The model choice and the interpretation of the parameters
	are discussed as well as the use of the identifiability concept for
	fixed partition models. The concept is generalized to "partial identifiability".},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Reboussin2001,
  author = {Reboussin, B. A. and Anthony, J. C.},
  title = {Latent class marginal regression models for modelling youthful drug
	involvement and its suspected influences},
  journal = {Statistics in Medicine},
  year = {2001},
  volume = {20},
  pages = {623-639},
  number = {4},
  abstract = {In longitudinal behavioural studies, it is common to have multiple
	categorical indicators for measuring a theoretical construct of interest.
	A latent class model is presented that accounts for the structure
	in a set of correlated, categorical variables measured at discrete
	time periods, drawing information from these Variables to form a
	smaller number of latent classes. The dependence of the resulting
	latent class model parameters on suspected factors over time is simultaneously
	modelled using a baseline-category logistic regression model. Estimation
	of the model parameters is achieved using an estimating equations
	procedure. A motivating example is provided from a longitudinal study
	of suspected linkages between monitoring or supervision by parents
	and the occurrence of drug use behaviours in an epidemiologic sample
	of school-attending youths. Copyright (C) 2001 John Wiley & Sons,
	Ltd.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Lin2002,
  author = {Lin, H. Q. and Turnbull, B. W. and McCulloch, C. E. and Slate, E.
	H.},
  title = {Latent class models for joint analysis of longitudinal biomarker
	and event process data: Application to longitudinal prostate-specific
	antigen readings and prostate cancer},
  journal = {Journal of the American Statistical Association},
  year = {2002},
  volume = {97},
  pages = {53-65},
  number = {457},
  abstract = {A retrospective substudy of the nutritional prevention of cancer (NPC)
	trials investigated the utility of longitudinally measured prostate-specific
	antigen (PSA) as a biomarker for subsequent onset of prostate cancer
	(PCa). Serial PSA levels were determined retrospectively from frozen
	blood samples that had been collected from all patients at successive
	clinic visits with the timing and the number of these visits highly
	variable. Diagnosis dates of all incident cases of PCa were recorded.
	Heterogeneity in PSA trajectories was observed that could not be
	fully explained by the usual linear mixed-effects model and measured
	covariates. Latent class models that incorporate both a longitudinal
	blomarker process and an event process offer a way to handle additional
	heterogeneity, to uncover distinct subpopulations. to incorporate
	correlated nonnormally distributed outcomes, and to classify individuals
	into risk classes, Our latent class joint model can aid the prediction
	of PCa probability given the longitudinal biomarker information available
	on an individual up to any date. The proposed model easily accommodates
	highly unbalanced longitudinal data and recurrent events. There are
	two levels of structure in the latent class joint model, First, the
	uncertainty of latent class membership is specified through a multinomial
	logistic model. Second, the class-specific marker trajectory and
	event process are specified parametrically and semiparametrically,
	under the assumption of conditional independence given the latent
	class membership. We use a likelihood approach to obtain parameter
	estimates via the EM algorithm. We fit the latent class joint model
	to the data from the NPC trials; four distinct subpopulations are
	identified that differ with regard to their PSA trajectories and
	risk for prostate cancer. Higher PSA level is significantly associated
	with increased risk of PCa, but appears to be conditionally independent
	once the latent classes are taken into account. Among the covariates,
	selenium supplementation and age at entry are statistically significant
	for various parts of the model. Assumptions-in particular the conditional
	independence between the longitudinal PSA blomarker and time to PCa
	diagnosis-are assessed.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Ng2003,
  author = {Ng, S. K. and McLachlan, G. J.},
  title = {An EM-based semi-parametric mixture model approach to the regression
	analysis of competing-risks data},
  journal = {Statistics in Medicine},
  year = {2003},
  volume = {22},
  pages = {1097-1111},
  number = {7},
  abstract = {We consider a mixture model approach to the regression analysis of
	competing-risks data. Attention is focused on inference concerning
	the effects of factors on both the probability of occurrence and
	the hazard rate conditional on each of the failure types. These two
	quantities are specified in the mixture model using the logistic
	model and the proportional hazards model, respectively. We propose
	a semi-parametric mixture method to estimate the logistic and regression
	coefficients jointly, whereby the component-baseline hazard functions
	are completely unspecified. Estimation is based on maximum likelihood
	on the basis of the full likelihood, implemented via an expectation-conditional
	maximization (ECM) algorithm. Simulation studies are performed to
	compare the performance of the proposed semi-parametric method with
	a fully parametric mixture approach. The results show that when the
	component-baseline hazard is monotonic increasing, the semi-parametric
	and fully parametric mixture approaches are comparable for mildly
	and moderately censored samples. When the component-baseline hazard
	is not monotonic increasing, the semi-parametric method consistently
	provides less biased estimates than a fully parametric approach and
	is comparable in efficiency in the estimation of the parameters for
	all levels of censoring. The methods are illustrated using a real
	data set of prostate cancer patients treated with different dosages
	of the drug diethylstilbestrol. Copyright (C) 2003 John Wiley Sons,
	Ltd.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.29}
}

@ARTICLE{Newton1994,
  author = {Newton, M. A. and Raftery, A. E. and Davison, A. C. and Bacha, M.
	and Celeux, G. and Carlin, B. P. and Clifford, P. and Lu, C. and
	Sherman, M. and Tanner, M. A. and Gelfand, A. E. and Mallick, B.
	K. and Gelman, A. and Grieve, A. P. and Kunsch, H. R. and Leonard,
	T. and Hsu, J. S. J. and Liu, J. S. and Rubin, D. B. and Lo, A. Y.
	and Louis, T. A. and Neal, R. M. and Owen, A. B. and Tu, D. S. and
	Gilks, W. R. and Roberts, G. and Sweeting, T. and Bates, D. and Ritter,
	G. and Worton, B. J. and Barnard, G. A. and Gibbens, R. and Silverman,
	B.},
  title = {Approximate Bayesian-inference with the weighted likelihood bootstrap},
  journal = {Journal of the Royal Statistical Society Series B-Methodological},
  year = {1994},
  volume = {56},
  pages = {3-48},
  number = {1},
  abstract = {We introduce the weighted likelihood bootstrap (WLB) as a way to simulate
	approximately from a posterior distribution. This method is often
	easy to implement, requiring only an algorithm for calculating the
	maximum likelihood estimator, such as iteratively reweighted least
	squares. In the generic weighting scheme, the WLB is first order
	correct under quite general conditions. Inaccuracies can be removed
	by using the WLB as a source of samples in the sampling-importance
	resampling (SIR) algorithm, which also allows incorporation of particular
	prior information. The SIR-adjusted WLB can be a competitive alternative
	to other integration methods in certain models. Asymptotic expansions
	elucidate the second-order properties of the WLB, which is a generalization
	of Rubin's Bayesian bootstrap. The calculation of approximate Bayes
	factors for model comparison is also considered. We note that, given
	a sample simulated from the posterior distribution, the required
	marginal likelihood may be simulation consistently estimated by the
	harmonic mean of the associated likelihood values; a modification
	of this estimator that avoids instability is also noted. These methods
	provide simple ways of calculating approximate Bayes factors and
	posterior model probabilities for a very wide class of models.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@ARTICLE{Mkhadri1997,
  author = {Mkhadri, A. and Celeux, G. and Nasroallah, A.},
  title = {Regularization in discriminant analysis: An overview},
  journal = {Computational Statistics \& Data Analysis},
  year = {1997},
  volume = {23},
  pages = {403-423},
  number = {3},
  abstract = {This paper presents an overview of regularized techniques in discriminant
	analysis. The case of continuous variables is treated first, and
	then the case of discrete variables. Three types of approaches are
	distinguished: combining standard methods, constraining models and
	Bayesian modelling. We include numerical experiments to assess the
	efficiency of regularized versions of predictive discrimination and
	to illustrate the superiority of regularization on variable subset
	selection in a small sample setting.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@ARTICLE{Li2006,
  author = {Li, T.},
  title = {A unified view on clustering binary data},
  journal = {Machine Learning},
  year = {2006},
  volume = {62},
  pages = {199-215},
  number = {3},
  abstract = {Clustering is the problem of identifying the distribution of patterns
	and intrinsic correlations in large data sets by partitioning the
	data points into similarity classes. This paper studies the problem
	of clustering binary data. Binary data have been occupying a special
	place in the domain of data analysis. A unified view of binary data
	clustering is presented by examining the connections among various
	clustering criteria. Experimental studies are conducted to empirically
	verify the relationships.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@ARTICLE{Eaton1989a,
  author = {Eaton, W. W. and Dryman, A. and Sorenson, A. and McCutcheon, A.},
  title = {DSM-III MAJOR DEPRESSIVE DISORDER IN THE COMMUNITY - A LATENT CLASS
	ANALYSIS OF DATA FROM THE NIMH EPIDEMIOLOGIC CATCHMENT-AREA PROGRAM},
  journal = {British Journal of Psychiatry},
  year = {1989},
  volume = {155},
  pages = {48-54},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@ARTICLE{Clifford1993,
  author = {Clifford, P. and Jennison, C. and Wakefield, J. and Phillips, D.
	and Frigessi, A. and Gray, A. J. and Lawson, A. and Forster, J. and
	Ramgopal, P. and Arslan, O. and Constable, P. D. L. and Kent, J.
	T. and Wolff, R. C. L. and Harding, E. F. and Middleton, R. and Diggle,
	P. J. and Aykroyd, R. G. and Berzuini, C. and Brewer, M. and Aitken,
	C. and Celeux, G. and Diebolt, J. and Critchley, F. and Diaconis,
	P. and Rosenthal, J. S. and Diebolt, J. and Robert, C. P. and Gelfand,
	A. E. and Lee, T. M. and Gelman, A. and Rubin, D. B. and Geman, D.
	and Geweke, J. and Geyer, C. J. and Gigli, A. and Givens, G. H. and
	Goodall, C. and Jonalasinio, G. D. and Grieve, A. P. and Han, X.
	L. and Kolassa, J. E. and Tanner, M. A. and Kooperberg, C. and Lewis,
	S. M. and Lin, S. and Thompson, E. A. and Litton, C. D. and Buck,
	C. E. and Liu, C. H. and Liu, J. and Mardia, K. V. and Marriott,
	J. M. and Moller, J. and Raftery, A. E. and Shephard, N. and Sinha,
	D. and Sokal, A. D. and Titterington, D. M. and Wilson, J. D. and
	York, J. and Madigan, D. and Smith, A. F. M. and Roberts, G. O. and
	Besag, J. and Green, P. J. and Gilks, W. R. and Clayton, D. G. and
	Spiegelhalter, D. J.},
  title = {Discussion on the meeting on the Gibbs sampler and other Markov chain
	Monte Carlo methods},
  journal = {Journal of the Royal Statistical Society Series B-Methodological},
  year = {1993},
  volume = {55},
  pages = {53-102},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@ARTICLE{Govaert2003,
  author = {Govaert, G. and Nadif, M.},
  title = {Clustering with block mixture models},
  journal = {Pattern Recognition},
  year = {2003},
  volume = {36},
  pages = {463-473},
  number = {2},
  abstract = {Basing cluster analysis on mixture models has become a classical and
	powerful approach. Until now, this approach, which allows to explain
	some classic clustering criteria such as the well-known k-means criteria
	and to propose general criteria, has been developed to classify a
	set of objects measured on a set of variables. But, for this kind
	of data, if most clustering procedures are designated to construct
	an optimal partition of objects or, sometimes, of variables, there
	exist others methods, named block clustering methods, which consider
	simultaneously the two sets and organize the data into homogeneous
	blocks. In this work, a new mixture model called block mixture model
	is proposed to take into account this situation. This model allows
	to embed simultaneous clustering of objects and variables in a mixture
	approach. We first consider this probabilistic model in a general
	context and we develop a new algorithm of simultaneous partitioning
	based on the CEM algorithm. Then, we focus on the case of binary
	data and we show that our approach allows us to extend a block clustering
	method, which had been proposed in this case, Simplicity. fast convergence
	and the possibility to process large data sets are the major advantages
	of the proposed approach. (C) 2002 Pattern Recognition Society. Published
	by Elsevier Science Ltd. All rights reserved.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@ARTICLE{Govaert1996,
  author = {Govaert, G. and Nadif, M.},
  title = {Comparison of the mixture and the classification maximum likelihood
	in cluster analysis with binary data},
  journal = {Computational Statistics \& Data Analysis},
  year = {1996},
  volume = {23},
  pages = {65-81},
  number = {1},
  abstract = {In this paper we propose to extend the comparison between the maximum
	likelihood and the classification maximum likelihood approaches for
	the Gaussian mixture (Ganesalingam, 1989; Celeux and Govaert, 1993)
	in the case of binary data, To this end, we use Bernoulli distribution
	mixtures. As with continuous data, two situations are discussed:
	first where mixing proportions are taken to be equal and secondly
	where they are unknown. The comparison realized with Monte-Carlo
	numerical experiments confirms the results obtained with continuous
	data. The choice of the approach depends on the size of the sample
	but assumptions about the mixing proportions are more important than
	the choice between the two approaches.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@INPROCEEDINGS{Celeux1991a,
  author = {Celeux, G. and Hebrail, G. and Mkhadri, A. and Suchard, M.},
  title = {Reduction of a large-scale and ill-conditioned statistical problem
	on textual data},
  booktitle = {5th International Symp on Applied Stochastic Models and Data Analysis},
  year = {1991},
  editor = {Gutierrez, R. Valderrama Mj},
  pages = {129-137},
  address = {Granada, Spain},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@ARTICLE{Celeux2001,
  author = {Celeux, G. and Chretien, S. and Forbes, F. and Mkhadri, A.},
  title = {A component-wise EM algorithm for mixtures},
  journal = {Journal of Computational and Graphical Statistics},
  year = {2001},
  volume = {10},
  pages = {697-712},
  number = {4},
  abstract = {Maximum likelihood estimation in finite mixture distributions is typically
	approached as an incomplete data problem to allow application of
	the expectation-maximization (EM) algorithm. In its general formulation,
	the EM algorithm involves the notion of a complete data space, in
	which the observed measurements and incomplete data are embedded.
	An advantage is that many difficult estimation problems are facilitated
	when viewed in this way. One drawback is that the simultaneous update
	used by standard EM requires overly informative complete data spaces,
	which leads to slow convergence in some situations. In the incomplete
	data context, it has been shown that the use of less informative
	complete data spaces, or equivalently smaller missing data spaces,
	can lead to faster convergence without sacrifying simplicity. However,
	in the mixture case, little progress has been made in speeding up
	EM. In this article we propose a component-wise EM for mixtures.
	It uses, at each iteration, the smallest admissible missing data
	space by intrinsically decoupling the parameter updates. Monotonicity
	is maintained, although the estimated proportions may not sum to
	one during the course of the iteration. However, we prove that the
	mixing proportions will satisfy this constraint upon convergence.
	Our proof of convergence relies on the interpretation of our procedure
	as a proximal point algorithm. For performance comparison, we consider
	standard EM as well as two other algorithms based on missing data
	space reduction, namely the SAGE and AECME algorithms. We provide
	adaptations of these general procedures to the mixture case. We also
	consider the ECME algorithm, which is not a data augmentation scheme
	but still aims at accelerating EM. Our numerical experiments illustrate
	the advantages of the component-wise EM algorithm relative to these
	other methods.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@ARTICLE{Celeux1996b,
  author = {Celeux, G. and Chauveau, D. and Diebolt, J.},
  title = {Stochastic versions of the EM algorithm: An experimental study in
	the mixture case},
  journal = {Journal of Statistical Computation and Simulation},
  year = {1996},
  volume = {55},
  pages = {287-314},
  number = {4},
  abstract = {We compare three different stochastic versions of the EM algorithm:
	The Stochastic EM algorithm (SEM), the ''Simulated Annealing'' EM
	algorithm (SAEM) and the Monte Carlo EM algorithm (MCEM). We focus
	particularly on the mixture of distributions problem. In this context,
	we investigate the practical behaviour of these algorithms through
	intensive Monte Carlo numerical simulations and a real data study.
	We show that, for some particular mixture situations, the SEM algorithm
	is almost always preferable to the EM and ''simulated annealing''
	versions SAEM and MCEM. For some severely overlapping mixtures, however,
	none of these algorithms can be confidently used. Then, SEM can be
	used as an efficient data exploratory tool for locating significant
	maxima of the likelihood function. In the real data case, we show
	that the SEM stationary distribution provides a contrasted view of
	the loglikelihood by emphasizing sensible maxima.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@ARTICLE{Biernacki2006,
  author = {Biernacki, C. and Celeux, G. and Govaert, G. and Langrognet, F.},
  title = {Model-based cluster and discriminant analysis with the MIXMOD software},
  journal = {Computational Statistics \& Data Analysis},
  year = {2006},
  volume = {51},
  pages = {587-600},
  number = {2},
  abstract = {The Mixture Modeling (MIXMOD) program fits mixture models to a given
	data set for the purposes of density estimation, clustering or discriminant
	analysis. A large variety of algorithms to estimate the mixture parameters
	are proposed (EM, Classification EM, Stochastic EM), and it is possible
	to combine these to yield different strategies for obtaining a sensible
	maximum for the likelihood (or complete-data likelihood) function.
	MIXMOD is currently intended to be used for multivariate Gaussian
	mixtures, and fourteen different Gaussian models can be distinguished
	according to different assumptions regarding the component variance
	matrix eigenvalue decomposition. Moreover, different information
	criteria for choosing a parsimonious model (the number of mixture
	components, for instance) are included, their suitability depending
	on the particular perspective (cluster analysis or discriminant analysis).
	Written in C++, MIXMOD is interfaced with SCILAB and MATLAB. The
	program, the statistical documentation and the user guide are available
	on the internet at the following address: http://www-math.univ-fcomte.fr/mixmod/index.php.
	(c) 2006 Elsevier B.V. All rights reserved.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@ARTICLE{Biernacki2003,
  author = {Biernacki, C. and Celeux, G. and Govaert, G.},
  title = {Choosing starting values for the EM algorithm for getting the highest
	likelihood in multivariate Gaussian mixture models},
  journal = {Computational Statistics \& Data Analysis},
  year = {2003},
  volume = {41},
  pages = {561-575},
  number = {3-4},
  abstract = {Simple methods to choose sensible starting values for the EM algorithm
	to get maximum likelihood parameter estimation in mixture models
	are compared. They are based on random initialization, using a classification
	EM algorithm (CEM), a Stochastic EM algorithm (SEM) or previous short
	runs of EM itself. Those initializations are included in a search/run/select
	strategy which can be compounded by repeating the three steps. They
	are compared in the context of multivariate Gaussian mixtures on
	the basis of numerical experiments on both simulated and real data
	sets in a target number of iterations. The main conclusions of those
	numerical experiments are the following. The simple random initialization
	which is probably the most employed way of initiating EM is often
	outperformed by strategies using CEM, SEM or shorts runs of EM before
	running EM. Also, it appears that compounding is generally profitable
	since using a single run of EM can often lead to suboptimal solutions.
	Otherwise, none of the experimental strategies can be regarded as
	the best one and it is difficult to characterize situations where
	a particular strategy can be expected to outperform the other ones.
	However, the strategy initiating EM with short runs of EM can be
	recommended. This strategy, which as far as we know was not used
	before the present study, has some advantages. It is simple, performs
	well in a lot of situations presupposing no particular form of the
	mixture to be fitted to the data and seems little sensitive to noisy
	data. (C) 2002 Elsevier Science B.V. All rights reserved.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@ARTICLE{Biernacki2000,
  author = {Biernacki, C. and Celeux, G. and Govaert, G.},
  title = {Assessing a mixture model for clustering with the integrated completed
	likelihood},
  journal = {Ieee Transactions on Pattern Analysis and Machine Intelligence},
  year = {2000},
  volume = {22},
  pages = {719-725},
  number = {7},
  abstract = {We propose assessing a mixture model in a cluster analysis setting
	with the integrated completed likelihood. With this purpose, the
	observed data are assigned to unknown clusters using a maximum a
	posteriori operator. Then, the Integrated Completed Likelihood (ICL)
	is approximated using an a` la Bayesian information criterion (BIC).
	Numerical experiments on simulated and real data of the resulting
	ICL criterion show that it performs well both for choosing a mixture
	model and a relevant number of clusters. In particular. ICL appears
	to be more robust than BIC to violation of some of the mixture model
	assumptions and it can select a number of clusters leading to a sensible
	partitioning of the data.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@ARTICLE{Sneath1962,
  author = {Sneath, P. H. A. and Sokal, R. R.},
  title = {NUMERICAL TAXONOMY},
  journal = {Nature},
  year = {1962},
  volume = {193},
  pages = {855-\&},
  number = {4818},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@ARTICLE{Robert1997,
  author = {Robert, C. P. and Aitkin, M. and Cox, D. R. and Stephens, M. and
	Polymenis, A. and Gilks, W. R. and Nobile, A. and Hodgson, M. and
	Ohagan, A. and Longford, N. T. and Dawid, A. P. and Atkinson, A.
	C. and Bernardo, J. M. and Besag, J. and Brooks, S. P. and Byers,
	S. and Raftery, A. and Celeux, G. and Cheng, R. C. H. and Liu, W.
	B. and Chien, Y. H. and George, E. I. and Cressie, N. and Huang,
	H. C. and Gruet, M. A. and Heath, S. C. and Jennison, C. and Lawson,
	A. B. and Clark, A. and McLachlan, G. and Peel, D. and Mengersen,
	K. and George, A. and Philippe, A. and Roeder, K. and Wasserman,
	L. and Schlattmann, P. and Bohning, D. and Titterington, D. M. and
	Tong, H. and West, M.},
  title = {On Bayesian analysis of mixtures with an unknown number of components
	- Discussion},
  journal = {Journal of the Royal Statistical Society Series B-Statistical Methodology},
  year = {1997},
  volume = {59},
  pages = {758-792},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@ARTICLE{Redner1984,
  author = {Redner, R. A. and Walker, H. F.},
  title = {Mixture densities, maximum likelihood and the EM algorithm},
  journal = {SIAM Review},
  year = {1984},
  volume = {26},
  pages = {195-237},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.30}
}

@ARTICLE{Qian2006,
  author = {Qian, Zhiguang and Shapiro, Alexander},
  title = {Simulation-based approach to estimation of latent variable models},
  journal = {Computational Statistics \& Data Analysis},
  year = {2006},
  volume = {51},
  pages = {1243-1259},
  number = {2},
  note = {0167-9473 doi: DOI: 10.1016/j.csda.2006.02.016},
  abstract = {We propose a simulation-based method for calculating maximum likelihood
	estimators in latent variable models. The proposed method integrates
	a recently developed sampling strategy, the so-called Sample Average
	Approximation method, to efficiently compute high quality solutions
	of the estimation problem. Theoretical and algorithmic properties
	of the method are discussed. A computational study, involving two
	numerical examples, is presented to highlight a significant improvement
	of the proposed approach over existing methods.},
  keywords = {Simulation-based inference Latent variable models Stochastic programming
	Sample average approximation},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.31}
}

@ARTICLE{Biernacki2006a,
  author = {Biernacki, Christophe and Celeux, Gilles and Govaert, GÈrard and
	Langrognet, Florent},
  title = {Model-based cluster and discriminant analysis with the MIXMOD software},
  journal = {Computational Statistics \& Data Analysis},
  year = {2006},
  volume = {51},
  pages = {587-600},
  number = {2},
  note = {0167-9473 doi: DOI: 10.1016/j.csda.2005.12.015},
  abstract = {The Mixture Modeling (MIXMOD) program fits mixture models to a given
	data set for the purposes of density estimation, clustering or discriminant
	analysis. A large variety of algorithms to estimate the mixture parameters
	are proposed (EM, Classification EM, Stochastic EM), and it is possible
	to combine these to yield different strategies for obtaining a sensible
	maximum for the likelihood (or complete-data likelihood) function.
	MIXMOD is currently intended to be used for multivariate Gaussian
	mixtures, and fourteen different Gaussian models can be distinguished
	according to different assumptions regarding the component variance
	matrix eigenvalue decomposition. Moreover, different information
	criteria for choosing a parsimonious model (the number of mixture
	components, for instance) are included, their suitability depending
	on the particular perspective (cluster analysis or discriminant analysis).
	Written in C++, MIXMOD is interfaced with SCILAB and MATLAB. The
	program, the statistical documentation and the user guide are available
	on the internet at the following address: http://www-math.univ-fcomte.fr/mixmod/index.php.},
  keywords = {Gaussian models EM-like algorithms Model selection},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.31}
}

@ARTICLE{Leisch2006,
  author = {Leisch, Friedrich},
  title = {A toolbox for {$K$}-centroids cluster analysis},
  journal = {Computational Statistics \& Data Analysis},
  year = {2006},
  volume = {51},
  pages = {526-544},
  number = {2},
  note = {DOI: 10.1016/j.csda.2005.10.006},
  abstract = {A methodological and computational framework for centroid-based partitioning
	cluster analysis using arbitrary distance or similarity measures
	is presented. The power of high-level statistical computing environments
	like R enables data analysts to easily try out various distance measures
	with only minimal programming effort. A new variant of centroid neighborhood
	graphs is introduced which gives insight into the relationships between
	adjacent clusters. Artificial examples and a case study from marketing
	research are used to demonstrate the influence of distances measures
	on partitions and usage of neighborhood graphs.},
  keywords = {Cluster analysis Distance measures R},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.31}
}

@ARTICLE{Lauro1996,
  author = {Lauro, Carlo},
  title = {Computational statistics or statistical computing, is that the question?},
  journal = {Computational Statistics \& Data Analysis},
  year = {1996},
  volume = {23},
  pages = {191-193},
  number = {1},
  note = {0167-9473 doi: DOI: 10.1016/0167-9473(96)88920-1},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.31}
}

@ARTICLE{Rousseeuw1996,
  author = {Rousseeuw, P. J. and Kaufman, L. and Trauwaert, E.},
  title = {Fuzzy clustering using scatter matrices},
  journal = {Computational Statistics \& Data Analysis},
  year = {1996},
  volume = {23},
  pages = {135-151},
  number = {1},
  note = {0167-9473 doi: DOI: 10.1016/S0167-9473(96)00026-6},
  abstract = {Starting from the well-known fuzzy k-means method, which was mainly
	intended for spherical clusters, several methods are considered which
	incorporate cluster-specific scatter matrices. This enables them
	to describe elliptical clusters with different orientation. The distinction
	between these methods lies in the way they deal with clusters of
	different volume, cardinality, and density. Some industrial examples
	show that different applications may lead to different goals and
	preferences, which affect the choice of the clustering method.},
  keywords = {Ellipsoidal clusters Fuzzy clustering Industrial applications Maximum
	likelihood SAND method},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.31}
}

@ARTICLE{Hartigan1996,
  author = {Hartigan, J. A.},
  title = {Recognition},
  journal = {Computational Statistics \& Data Analysis},
  year = {1996},
  volume = {23},
  pages = {97-103},
  number = {1},
  note = {0167-9473 doi: DOI: 10.1016/S0167-9473(96)00023-0},
  abstract = {In making a prediction, we divide our attention between objects presently
	perceived and previously experienced objects. The present objects
	are recognised as similar to objects previously experienced, and
	the qualities remembered from previous examination are predicted
	for the present objects. Prediction is fallible, in that we may make
	errors in recognising the present object, or in past observations
	of the experienced object, or in assigning qualities to the present
	object which hold for the experienced objects similar to the present
	object, but not for the present object. We use probability to quantify
	these errors. We use classification in organizing our experiences,
	and in recognising present objects as being similar to some species
	of experienced objects.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.31}
}

@ARTICLE{VanCutsem1996,
  author = {Van Cutsem, Bernard},
  title = {Combinatorial structures and structures for classification},
  journal = {Computational Statistics \& Data Analysis},
  year = {1996},
  volume = {23},
  pages = {169-188},
  number = {1},
  note = {0167-9473 doi: DOI: 10.1016/S0167-9473(96)00028-X},
  abstract = {The problem of random generation of structures of classification,
	such as partitions or various kinds of hierarchies, is approached
	in two different ways. On the one hand, structures of classification
	are considered as particular combinatorial structures and then they
	can be randomly generated using the calculus introduced by Flajolet,
	Zimmermann and Van Cutsem (1993). On the other hand, some Markov
	or renewal process models can be introduced to describe some of these
	structures and can be used to generate such random structures, according
	to some ideas of Van Cutsem and Ycart (1993). Some asymptotics properties
	allow one to simplify the procedures. These last results are presented
	in the example of stratified hierarchies. It is clear that these
	techniques are mainly useful for the problem of the classifiability
	of data.},
  keywords = {Combinatorial structures operations and specifications, uniform random
	generation Structures of classification, partitions, hierarchies
	Markov chains},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.31}
}

@ARTICLE{Hardy1996,
  author = {Hardy, AndrÈ},
  title = {On the number of clusters},
  journal = {Computational Statistics \& Data Analysis},
  year = {1996},
  volume = {23},
  pages = {83-96},
  number = {1},
  note = {0167-9473 doi: DOI: 10.1016/S0167-9473(96)00022-9},
  abstract = {A large number of classification and clustering methods for defining
	and calculating optimal or well-suited partitions for data sets are
	available. Perhaps the most difficult problem facing the user of
	cluster analysis techniques in practice is the objective assessment
	of the stability and validity of the clusters found by the numerical
	technique used. The problem of determining the "true" number of clusters
	has been called the fundamental problem of cluster validity. The
	aim of this paper is to compare three methods based on the hypervolume
	criterion with other well-known methods. To illustrate and compare
	their behaviour, these procedures for determining the number of clusters
	are applied to artificially constructed bivariate data containing
	various types of structure. To provide a variety of solutions six
	clustering methods are used. We finally conclude by pointing out
	the performance of each method and by giving some recommendations
	to help potential users of these techniques.},
  keywords = {Cluster analysis Hypervolume criterion Number of clusters},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.31}
}

@ARTICLE{Hartigan1996a,
  author = {Hartigan, J. A.},
  title = {Introduction: {C}lassification, probability and statistics},
  journal = {Computational Statistics \& Data Analysis},
  year = {1996},
  volume = {23},
  pages = {3-4},
  number = {1},
  note = {0167-9473 doi: DOI: 10.1016/S0167-9473(96)00017-5},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.31}
}

@ARTICLE{Rasson1996,
  author = {Rasson, J. P. and Rousseeuw, P. J.},
  title = {Preface},
  journal = {Computational Statistics \& Data Analysis},
  year = {1996},
  volume = {23},
  pages = {1-1},
  number = {1},
  note = {0167-9473 doi: DOI: 10.1016/S0167-9473(96)90279-0},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.31}
}

@BOOK{Hartigan1975,
  title = {Clustering algorithms},
  publisher = {Wiley},
  year = {1975},
  author = {Hartigan, John A.},
  address = {New York,},
  keywords = {Cluster analysis. Cluster analysis Data processing.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.31}
}

@INBOOK{McCutcheon2002,
  chapter = {Basic concepts and procedures in single- and multiple-group latent
	class analysis},
  pages = {56-85},
  title = {Applied latent class analysis},
  publisher = {Cambridge University Press},
  year = {2002},
  editor = {Hagenaars, Jacques A. and McCutcheon, Allan L.},
  author = {McCutcheon, Allan L.},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.31}
}

@INBOOK{Vermunt2002,
  chapter = {Latent class cluster analysis},
  pages = {89-106},
  title = {Applied latent class analysis},
  publisher = {Cambridge University Press},
  year = {2002},
  editor = {Hagenaars, Jacques A. and McCutcheon, Allan L.},
  author = {Vermunt, Jeroen K. and Magidson, Jay},
  owner = {rebeccaspeckman},
  timestamp = {2009.01.31}
}

@ARTICLE{Goodman1974,
  author = {Goodman, Leo A.},
  title = {The Analysis of Systems of Qualitative Variables When Some of the
	Variables Are Unobservable. Part I-A Modified Latent Structure Approach},
  journal = {The American Journal of Sociology},
  year = {1974},
  volume = {79},
  pages = {1179-1259},
  number = {5},
  note = {00029602 ArticleType: primary_article / Full publication date: Mar.,
	1974 / Copyright © 1974 The University of Chicago Press},
  abstract = {This article presents methods for analyzing the relationships among
	a set of qualitative variables when some of these variables are specified
	manifest (i.e., observed) variables and others are latent (i.e.,
	unobserved or unobservable) variables. We shall show how to estimate
	the magnitude of the various effects represented in pathdiagram models
	that include both the manifestand latent variables, and also how
	to test whether this kind of path-diagram model is congruent with
	the observed data. These methods can be applied in order to analyze
	data obtained in various kinds of surveys (including panel studies),
	and also in order to construct tests and indices for purposes of
	measurement and prediction. To illustrate their wide applicability
	and flexibility, we shall use these methods to reanalyze several
	different sets of data which were analyzed earlier by Coleman (1964),
	Lazarsfeld (1948, 1970), Goodman (1973a), and others. Except for
	some related conclusions in Goodman (1973a), the methods introduced
	herein lead to conclusions that are very different from those presented
	by the other researchers who had analyzed these data earlier.},
  owner = {rebeccaspeckman},
  timestamp = {2009.02.01}
}

@ARTICLE{Clogg1981,
  author = {Clogg, Clifford C.},
  title = {Latent Structure Models of Mobility},
  journal = {The American Journal of Sociology},
  year = {1981},
  volume = {86},
  pages = {836-868},
  number = {4},
  note = {00029602 ArticleType: primary_article / Full publication date: Jan.,
	1981 / Copyright © 1981 The University of Chicago Press},
  abstract = {This paper proposes several latent structure models for the analysis
	of mobility tables and examines the relationship of these to some
	earlier mobility models (e.g., the "perfect" and the "quasi-perfect"
	mobility models). Data from the classic Danish (5 X 5) and British
	(5 X 5 and 8 X 8) mobility tables are used to illustrate the utility
	of these methods in comparative analysis. A model designated as a
	quasi-latent structure is suggested as a plausible rendering of the
	structure of mobility for each set of data, and this model is used
	to derive various kinds of substantive inferences.},
  owner = {rebeccaspeckman},
  timestamp = {2009.02.03}
}

@ARTICLE{Baulieu1989,
  author = {Baulieu, F. B.},
  title = {A CLASSIFICATION OF PRESENCE ABSENCE BASED DISSIMILARITY COEFFICIENTS},
  journal = {Journal of Classification},
  year = {1989},
  volume = {6},
  pages = {233-246},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.02.27}
}

@ARTICLE{Baulieu1997,
  author = {Baulieu, F. B.},
  title = {Two variant axiom systems for presence/absence based dissimilarity
	coefficients},
  journal = {Journal of Classification},
  year = {1997},
  volume = {14},
  pages = {159-170},
  number = {1},
  abstract = {The axiomatic treatment of presence/absence based dissimilarity coefficients
	presented by Baulieu (1989) is extended. Independence of the axiom
	sets is established. Global order equivalence is examined as well
	as its relation to the axiom sets for two variations of the model.
	It is shown that if paired absences are to be ignored in the calculation
	of dissimilarity values, then one variant model admits only a single
	dissimilarity coefficient (up to global order equivalence): (a +
	b)/(a + b + c).},
  owner = {rebeccaspeckman},
  timestamp = {2009.02.27}
}

@CONFERENCE{MacQueen1967,
  author = {MacQueen},
  title = {Some methods for classification and analysis of multivariate observations.},
  booktitle = {Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics
	and Probability},
  year = {1967},
  pages = {281-297},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@INPROCEEDINGS{Guolo2006,
  author = {Guolo, A. and Brazzale, A. R.},
  title = {Empirical evaluation of measurement error correction techniques for
	matched case-control studies},
  booktitle = {ISEE/ISEA 2006 Conference},
  year = {2006},
  pages = {S437-S437},
  address = {Paris, FRANCE},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@ARTICLE{Brusco2004,
  author = {Brusco, M. J.},
  title = {Clustering binary data in the presence of masking variables},
  journal = {Psychological Methods},
  year = {2004},
  volume = {9},
  pages = {510-523},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@BOOK{Hubert2006,
  title = {The structural representation of proximity matrices with MATLAB},
  publisher = {Society for Industrial and Applied Mathematics ; American Statistical
	Association},
  year = {2006},
  author = {Hubert, Lawrence J. and Arabie, Phipps and Meulman, Jacqueline},
  series = {ASA-SIAM series on statistics and applied probability},
  address = {Philadelphia, Pa. Alexandria, Va.},
  keywords = {Proximity matrices. Functions. Representations of graphs.},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@BOOK{Arabie1996a,
  title = {Clustering and classification},
  publisher = {World Scientific},
  year = {1996},
  author = {Arabie, Phipps and Hubert, Lawrence J. and Soete, Geert de},
  address = {Singapore ; River Edge, NJ},
  keywords = {Cluster analysis. Discriminant analysis.},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@INBOOK{Arabie1996,
  chapter = {An overview of combinatorial data analysis},
  pages = {5--63},
  title = {Clustering and classification},
  publisher = {World Scientific},
  year = {1996},
  editor = {Arabie, Phipps and Hubert, Lawrence J. and Soete, Geert de},
  author = {Arabie, Phipps and Hubert, Lawrence J.},
  address = {Singapore ; River Edge, NJ},
  keywords = {Cluster analysis. Discriminant analysis.},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@ARTICLE{Sheng2006,
  author = {Sheng, W. G. and Liu, X. H.},
  title = {A genetic k-medoids clustering algorithm},
  journal = {Journal of Heuristics},
  year = {2006},
  volume = {12},
  pages = {447-466},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@ARTICLE{Xu2005,
  author = {Xu, R. and Wunsch, D.},
  title = {Survey of clustering algorithms},
  journal = {Ieee Transactions on Neural Networks},
  year = {2005},
  volume = {16},
  pages = {645-678},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@INPROCEEDINGS{Chiang2004,
  author = {Chiang, C. S. and Chu, S. C. and Hsin, Y. C. and Wang, M. H.},
  title = {Genetic distance measure for K-modes algorithm},
  booktitle = {International Workshop on Fuzzy Systems and Innovational Computing},
  year = {2004},
  pages = {33-40},
  address = {Kitalyushu, JAPAN},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@INPROCEEDINGS{Cerioli2006,
  author = {Cerioli, A. and Riani, M. and Atkinson, A. C.},
  title = {Robust classification with categorical variables},
  booktitle = {17th Symposium on Computational Statistics (COMSTAT 2006)},
  year = {2006},
  editor = {Rizzi, A. and Vichi, M.},
  pages = {507-519},
  address = {Rome, ITALY},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@ARTICLE{Vinod1969,
  author = {Vinod, H. D.},
  title = {INTEGER PROGRAMMING AND THEORY OF GROUPING},
  journal = {Journal of the American Statistical Association},
  year = {1969},
  volume = {64},
  pages = {506-\&},
  number = {326},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@ARTICLE{Massart1983,
  author = {Massart, D. L. and Plastria, F. and Kaufman, L.},
  title = {NON-HIERARCHICAL CLUSTERING WITH MASLOC},
  journal = {Pattern Recognition},
  year = {1983},
  volume = {16},
  pages = {507-516},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@ARTICLE{Abaya1984,
  author = {Abaya, E. F. and Wise, G. L.},
  title = {CONVERGENCE OF VECTOR QUANTIZERS WITH APPLICATIONS TO OPTIMAL QUANTIZATION},
  journal = {Siam Journal on Applied Mathematics},
  year = {1984},
  volume = {44},
  pages = {183-189},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@ARTICLE{Chaturvedi2001,
  author = {Chaturvedi, A. and Green, P. E. and Carroll, J. D.},
  title = {K-modes clustering},
  journal = {Journal of Classification},
  year = {2001},
  volume = {18},
  pages = {35-55},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@ARTICLE{Huang2003,
  author = {Huang, Z. X. and Ng, M. K.},
  title = {A note on $K$-modes clustering},
  journal = {Journal of Classification},
  year = {2003},
  volume = {20},
  pages = {257-261},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@INPROCEEDINGS{Couto2005,
  author = {Couto, J.},
  title = {Kernel K-means for categorical data},
  booktitle = {6th International Symposium on Intelligent Data Analysis},
  year = {2005},
  editor = {Famili, A. F. and Kok, J. N. and Pena, J. M. and Siebes, A. and Feelders,
	A.},
  pages = {46-56},
  address = {Madrid, SPAIN},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@ARTICLE{Estivill-Castro2004,
  author = {Estivill-Castro, V. and Yang, J.},
  title = {Fast and robust general purpose clustering algorithms},
  journal = {Data Mining and Knowledge Discovery},
  year = {2004},
  volume = {8},
  pages = {127-150},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.03}
}

@INCOLLECTION{Bock2007,
  author = {Bock, Hans-Hermann},
  title = {A {H}istory of $k$-{M}eans {A}lgorithms},
  booktitle = {Selected Contributions in Data Analysis and Classification},
  publisher = {Springer Berlin Heidelberg},
  year = {2007},
  series = {Studies in Classification, Data Analysis, and Knowledge Organization},
  pages = {161--172},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.04}
}

@BOOK{Xu2009,
  title = {Clustering},
  publisher = {IEEE Press},
  year = {2009},
  author = {Xu, Rui and Wunsch, Donald C.},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.04}
}

@BOOK{Spath1980,
  title = {Cluster analysis algorithms for data reduction and classification
	of objects},
  publisher = {E. Horwood ; Halsted Press},
  year = {1980},
  author = {Sp\"{a}th, Helmuth},
  series = {Computers and their applications},
  address = {Chichester, Eng. New York},
  keywords = {Cluster analysis Data processing. FORTRAN (Computer program language)},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.04}
}

@BOOK{Kogan2007,
  title = {Introduction to clustering large and high-dimensional data},
  publisher = {Cambridge University Press},
  year = {2007},
  author = {Kogan, Jacob},
  address = {Cambridge ; New York},
  keywords = {Cluster analysis Data processing. Cluster analysis Computer programs.
	Computer algorithms. Dimensional analysis Data processing. Dimensional
	analysis Computer programs.},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.04}
}

@BOOK{Gan2007,
  title = {Data clustering : theory, algorithms, and applications},
  publisher = {SIAM American Statistical Association},
  year = {2007},
  author = {Gan, Guojun and Ma, Chaoqun and Wu, Jianhong},
  address = {Philadelphia, Pa. Alexandria, Va.},
  keywords = {Cluster analysis. Cluster analysis Data processing.},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.04}
}

@BOOK{Anderberg1973,
  title = {Cluster analysis for applications},
  publisher = {Academic Press},
  year = {1973},
  author = {Anderberg, Michael R.},
  address = {New York,},
  keywords = {Cluster analysis.},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.04}
}

@BOOK{Spath1985,
  title = {Cluster dissection and analysis : {T}heory, FORTRAN programs, examples},
  publisher = {Horwood ; Halsted Press [distributor]},
  year = {1985},
  author = {Sp\"{a}th, Helmuth},
  address = {Chichester New York},
  keywords = {Cluster analysis Data processing.},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.04}
}

@INPROCEEDINGS{Ordonez2003,
  author = {Ordonez, Carlos},
  title = {Clustering binary data streams with K-means},
  booktitle = {DMKD '03: Proceedings of the 8th ACM SIGMOD workshop on Research
	issues in data mining and knowledge discovery},
  year = {2003},
  pages = {12--19},
  address = {New York, NY, USA},
  publisher = {ACM},
  location = {San Diego, California},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.04}
}

@ARTICLE{Janowitz1979,
  author = {Janowitz, M. F.},
  title = {MONOTONE EQUIVARIANT CLUSTER METHODS},
  journal = {Siam Journal on Applied Mathematics},
  year = {1979},
  volume = {37},
  pages = {148-165},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.06}
}

@ARTICLE{Janowitz1979a,
  author = {Janowitz, M. F.},
  title = {PRESERVATION OF GLOBAL ORDER EQUIVALENCE},
  journal = {Journal of Mathematical Psychology},
  year = {1979},
  volume = {20},
  pages = {78-88},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.06}
}

@ARTICLE{Janowitz1993,
  author = {Janowitz, M. F. and Stinebrickner, R.},
  title = {PRESERVATION OF WEAK ORDER EQUIVALENCE},
  journal = {Mathematical Social Sciences},
  year = {1993},
  volume = {25},
  pages = {181-197},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.06}
}

@ARTICLE{Emrich1991,
  author = {Emrich, L. J. and Piedmonte, M. R.},
  title = {A method for generating high-dimensional multivariate binary variates},
  journal = {American Statistician},
  year = {1991},
  volume = {45},
  pages = {302-304},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.06}
}

@INCOLLECTION{Kaufman1987,
  author = {Kaufman, Leonard and Rousseeuw, Peter J.},
  title = {Clustering by means of medoids.},
  booktitle = {Statistical data analysis based on the L b1 s-norm and related methods},
  publisher = {North-Holland ; Sole distributors for the U.S.A. and Canada, Elsevier
	Science Pub. Co.},
  year = {1987},
  editor = {Dodge, Yadolah},
  chapter = {Clustering by means of medoids},
  pages = {405--416},
  address = {Amsterdam ; New York New York, N.Y., U.S.A.},
  keywords = {Least absolute deviations (Statistics) Congresses. Mathematical statistics
	Congresses.},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.07}
}

@BOOK{Kaufman2005,
  title = {Finding groups in data : an introduction to cluster analysis},
  publisher = {Wiley},
  year = {2005},
  author = {Kaufman, Leonard and Rousseeuw, Peter J.},
  series = {Wiley series in probability and mathematical statistics},
  address = {Hoboken, N.J.},
  keywords = {Cluster analysis.},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.07}
}

@BOOK{Kaufman1990,
  title = {Finding groups in data : an introduction to cluster analysis},
  publisher = {Wiley},
  year = {1990},
  author = {Kaufman, Leonard and Rousseeuw, Peter J.},
  series = {Wiley series in probability and mathematical statistics. Applied
	probability and statistics,},
  address = {New York},
  keywords = {Cluster analysis.},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.07}
}

@ARTICLE{Ng2002,
  author = {Ng, R. T. and Han, J. W.},
  title = {CLARANS: A method for clustering objects for spatial data mining},
  journal = {Ieee Transactions on Knowledge and Data Engineering},
  year = {2002},
  volume = {14},
  pages = {1003-1016},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.07}
}

@ARTICLE{VanderLaan2003,
  author = {Van der Laan, M. J. and Pollard, K. S. and Bryan, J.},
  title = {A new partitioning around medoids algorithm},
  journal = {Journal of Statistical Computation and Simulation},
  year = {2003},
  volume = {73},
  pages = {575-584},
  number = {8},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.07}
}

@ARTICLE{Park2009,
  author = {Park, H. S. and Jun, C. H.},
  title = {A simple and fast algorithm for K-medoids clustering},
  journal = {Expert Systems with Applications},
  year = {2009},
  volume = {36},
  pages = {3336-3341},
  number = {2},
  note = {Part 2},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.07}
}

@ARTICLE{Selim1984,
  author = {Selim, S. Z. and Ismail, M. A.},
  title = {{$K$}-means-type algorithms: a generalized convergence theorem and
	characterization of local optimality},
  journal = {{IEEE} Transactions on Pattern Analysis and Machine Intelligence},
  year = {1984},
  volume = {6},
  pages = {81-87},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.07}
}

@ARTICLE{Bobrowski1991,
  author = {Bobrowski, L. and Bezdek, J. C.},
  title = {C-MEANS CLUSTERING WITH THE L1 AND L-INFINITY NORMS},
  journal = {Ieee Transactions on Systems Man and Cybernetics},
  year = {1991},
  volume = {21},
  pages = {545-554},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.07}
}

@INPROCEEDINGS{Leeser2000,
  author = {Leeser, M. and Theiler, J. and Estlick, M. and Szymanski, J. J.},
  title = {Design tradeoffs in a hardware implementation of the K-means clustering
	algorithm},
  booktitle = {1st IEEE Sensor Array and Multichannel Signal Processing Workshop
	(SAM 2000)},
  year = {2000},
  pages = {520-524},
  address = {Cambridge, Ma},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.07}
}

@ARTICLE{Spath1976,
  author = {Spath, H.},
  title = {ALGORITHM .30. L-1 CLUSTER-ANALYSIS},
  journal = {Computing},
  year = {1976},
  volume = {16},
  pages = {379-387},
  number = {4},
  note = {ISI Document Delivery No.: BT480 Times Cited: 1 Cited Reference Count:
	3},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.08}
}

@ARTICLE{Hamming1950,
  author = {Hamming, R. W.},
  title = {ERROR DETECTING AND ERROR CORRECTING CODES},
  journal = {Bell System Technical Journal},
  year = {1950},
  volume = {29},
  pages = {147-160},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.08}
}

@ARTICLE{Jancey1966,
  author = {Jancey, R. C.},
  title = {MULTIDIMENSIONAL GROUP ANALYSIS},
  journal = {Australian Journal of Botany},
  year = {1966},
  volume = {14},
  pages = {127-\&},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.08}
}

@ARTICLE{Ralambondrainy1995,
  author = {Ralambondrainy, H.},
  title = {A CONCEPTUAL VERSION OF THE K-MEANS ALGORITHM},
  journal = {Pattern Recognition Letters},
  year = {1995},
  volume = {16},
  pages = {1147-1157},
  number = {11},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.08}
}

@ARTICLE{Davies1979,
  author = {Davies, D. L. and Bouldin, D. W.},
  title = {A cluster separation measure},
  journal = {Ieee Transactions on Pattern Analysis and Machine Intelligence},
  year = {1979},
  volume = {1},
  pages = {224-227},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.13}
}

@INPROCEEDINGS{Halkidi2001,
  author = {Halkidi, M. and Batistakis, Y. and Vazirgiannis, M.},
  title = {On clustering validation techniques},
  booktitle = {13th International Conference on Scientific and Statistical Database
	Management (SSDBM 2001)},
  year = {2001},
  pages = {107-145},
  address = {Fairfax, Virginia},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.13}
}

@ARTICLE{Halkidi2002,
  author = {Halkidi, M. and Batistakis, Y. and Vazirgiannis, M.},
  title = {Cluster validity methods: Part I},
  journal = {Sigmod Record},
  year = {2002},
  volume = {31},
  pages = {40-45},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.13}
}

@ARTICLE{Ratkowsky1978,
  author = {Ratkowsky, D. A. and Lance, G. N.},
  title = {A criterion for determining the number of groups in a classification},
  journal = {Australian Computer Journal},
  year = {1978},
  volume = {10},
  pages = {115-117},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.13}
}

@ARTICLE{Hill1980,
  author = {Hill, R. S.},
  title = {A STOPPING RULE FOR PARTITIONING DENDROGRAMS},
  journal = {Botanical Gazette},
  year = {1980},
  volume = {141},
  pages = {321-324},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.13}
}

@INPROCEEDINGS{Xu1997,
  author = {Xu, L.},
  title = {Bayesian Ying-Yang machine, clustering and number of clusters},
  booktitle = {Pattern Recognition in Practice V Conference},
  year = {1997},
  pages = {1167-1178},
  address = {Vlieland, Netherlands},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.13}
}

@ARTICLE{Carmichael1968,
  author = {Carmichael, J. W. and George, J. Alan and Julius, R. S.},
  title = {Finding Natural Clusters},
  journal = {Systematic Zoology},
  year = {1968},
  volume = {17},
  pages = {144--150},
  number = {2},
  abstract = {On being presented with a set of points plotted on a plane space,
	most observers will agree that the set contains clusters of points
	if the distribution of the points meets certain conditions. We postulate
	that these conditions are: (1) that there are continuous, relatively
	densely populated regions of the space, and (2) that these are surrounded
	by continuous, relatively empty regions of the space. Using this
	definition for natural clusters we developed an algorithm for finding
	such clusters from the relative interpoint distances within a set.
	Starting with the closest pair of points, additional cluster members
	are sought by a minimum stepping stone strategy and accepted or rejected
	on the basis of parameters which indicate a discontinuity of closeness
	(i.e. a continuous, relatively empty space surrounding the points
	already admitted to the cluster) The procedure is repeated at a number
	of levels of resolution to give an overall view of the population.
	The resolution levels appropriate for a given set of points are determined
	from an analysis of the interpoint distances. Since the procedure
	uses only the relative distances between the points, it can be applied
	to points specified in any number of dimensions.},
  jstor_formatteddate = {Jun., 1968},
  owner = {rebeccaspeckman},
  publisher = {Taylor \& Francis, Ltd. for the Society of Systematic Biologists},
  timestamp = {2009.03.13}
}

@ARTICLE{Carmichael1969,
  author = {Carmichael, J. W. and Sneath, P. H. A.},
  title = {Taxometric Maps},
  journal = {Systematic Zoology},
  year = {1969},
  volume = {18},
  pages = {402--415},
  number = {4},
  abstract = {An n x t table, recording the results of n different tests or observations
	on each of t items, contains information which may be used to classify
	the items (OTU's) according to their similarities and differences.
	We equate similarity with relative closeness, and difference with
	relative distance. Relative closeness and relative distance may be
	thought of as complementary measures of proximity. The process of
	converting an n \times t table of test results into a t \times t
	table of proximities between the pairs of items is called proximity
	analysis. The process of finding any isolated subsets of similar
	items is called cluster analysis. Taxometric methods are viewed as
	approaches to compressing the information in an n \times t table
	into a graphic, best-fit model for the display of proximity and cluster
	relations among OTU's. Three general approaches are followed. 1)
	compression before proximity analysis, 2) compression as an intermediate
	step between proximity and cluster analysis, and 3) compression after
	cluster analysis. The utility of these approaches is compared. Taxometric
	maps are introduced as a means to display proximity and cluster relations.
	A procedure is given for preparing taxometric maps from cluster analysis
	results, and two examples are presented for illustration. An appendix
	on similarity and proximity equates three independently derived similarity
	coefficients, which are shown to be derivable from a city-block metric
	proximity measure.},
  jstor_formatteddate = {Dec., 1969},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.13}
}

@ARTICLE{Jain1999,
  author = {Jain, A. K. and Murty, M. N. and Flynn, P. J.},
  title = {Data clustering: A review},
  journal = {Acm Computing Surveys},
  year = {1999},
  volume = {31},
  pages = {264-323},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.16}
}

@ARTICLE{Grabmeier2002,
  author = {Grabmeier, J. and Rudolph, A.},
  title = {Techniques of cluster algorithms in data mining},
  journal = {Data Mining and Knowledge Discovery},
  year = {2002},
  volume = {6},
  pages = {303-360},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.16}
}

@ARTICLE{Gyllenberg1997,
  author = {Gyllenberg, M. and Koski, T. and Verlaan, M.},
  title = {Classification of binary vectors by stochastic complexity},
  journal = {Journal of Multivariate Analysis},
  year = {1997},
  volume = {63},
  pages = {47-72},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.16}
}

@ARTICLE{Lee1997,
  author = {Lee, A. J.},
  title = {Some simple methods for generating correlated categorical variates},
  journal = {Computational Statistics \& Data Analysis},
  year = {1997},
  volume = {26},
  pages = {133-148},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.16}
}

@ARTICLE{Franti2000,
  author = {Franti, P. and Gyllenberg, H. G. and Gyllenberg, M. and Kivijarvi,
	J. and Koski, T. and Lund, T. and Nevalainen, O.},
  title = {Minimizing stochastic complexity using local search and GLA with
	applications to classification of bacteria},
  journal = {Biosystems},
  year = {2000},
  volume = {57},
  pages = {37-48},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.16}
}

@ARTICLE{Kang2001,
  author = {Kang, S. H. and Jung, S. H.},
  title = {Generating correlated binary variables with complete specification
	of the joint distribution},
  journal = {Biometrical Journal},
  year = {2001},
  volume = {43},
  pages = {263-269},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.16}
}

@ARTICLE{Al-Osh2001,
  author = {Al-Osh, M. A. and Lee, S. J.},
  title = {A simple approach for generating correlated binary variates},
  journal = {Journal of Statistical Computation and Simulation},
  year = {2001},
  volume = {70},
  pages = {231-255},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.16}
}

@ARTICLE{Qaqish2003,
  author = {Qaqish, B. F.},
  title = {A family of multivariate binary distributions for simulating correlated
	binary variables with specified marginal means and correlations},
  journal = {Biometrika},
  year = {2003},
  volume = {90},
  pages = {455-463},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.16}
}

@ARTICLE{Demirtas2007,
  author = {Demirtas, H.},
  title = {The design of simulation studies in medical statistics},
  journal = {Statistics in Medicine},
  year = {2007},
  volume = {26},
  pages = {3818-3821},
  number = {20},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.16}
}

@ARTICLE{Demirtas2006,
  author = {Demirtas, H.},
  title = {A method for multivariate ordinal data generation given marginal
	distributions and correlations},
  journal = {Journal of Statistical Computation and Simulation},
  year = {2006},
  volume = {76},
  pages = {1017-1025},
  number = {11},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.16}
}

@ARTICLE{Farrell2008,
  author = {Farrell, P. J. and Rogers-Stewart, K.},
  title = {Methods for generating longitudinally correlated binary data},
  journal = {International Statistical Review},
  year = {2008},
  volume = {76},
  pages = {28-38},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.16}
}

@ARTICLE{Lee1993,
  author = {Lee, A. J.},
  title = {GENERATING RANDOM BINARY DEVIATES HAVING FIXED MARGINAL DISTRIBUTIONS
	AND SPECIFIED DEGREES OF ASSOCIATION},
  journal = {American Statistician},
  year = {1993},
  volume = {47},
  pages = {209-215},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.16}
}

@ARTICLE{Goodman1974a,
  author = {Goodman, L. A.},
  title = {EXPLORATORY LATENT STRUCTURE-ANALYSIS USING BOTH IDENTIFIABLE AND
	UNIDENTIFIABLE MODELS},
  journal = {Biometrika},
  year = {1974},
  volume = {61},
  pages = {215-231},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.16}
}

@INPROCEEDINGS{Birkelund1995,
  author = {Birkelund, G. E. and Goodman, L. A. and Rose, D.},
  title = {The latent structure of job characteristics of men and women},
  booktitle = {Biannual Meeting of the Research-Committee-on-Social-Stratification,
	International-Sociological-Association},
  year = {1995},
  pages = {80-113},
  address = {Zurich, Switzerland},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.16}
}

@INPROCEEDINGS{Burton2005,
  author = {Burton, A. and Altman, D. G. and Royston, P. and Holder, R. L.},
  title = {The design of simulation studies in medical statistics},
  booktitle = {26th Annual Meeting of the International-Society-for-Clinical-Biostatistics
	(ISCB26)},
  year = {2005},
  pages = {4279-4292},
  address = {Szeged, HUNGARY},
  abstract = {Simulation studies use computer intensive procedures to assess the
	performance of a variety of statistical methods in relation to a
	known truth. Such evaluation cannot be achieved with studies of real
	data alone. Designing high-quality simulations that reflect the complex
	situations seen in practice, such as in prognostic factors studies,
	is not a simple process. Unfortunately, very few published simulation
	studies provide sufficient details to allow readers to understand
	fully all the processes required to design a simulation study. When
	planning a simulation study, it is recommended that a detailed protocol
	be produced, giving full details of how the study will be performed,
	analysed and reported. This paper details the important considerations
	necessary when designing any simulation study, including defining
	specific objectives of the study, determining the procedures for
	generating the data sets and the number of simulations to perform.
	A checklist highlighting the important considerations when designing
	a simulation study is provided. A small review of the literature
	identifies the current practices within published simulation studies.
	Copyright (c) 2006 John Wiley & Sons, Ltd.},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.16}
}

@ARTICLE{Rousseeuw1987,
  author = {Rousseeuw, P. J.},
  title = {SILHOUETTES - A GRAPHICAL AID TO THE INTERPRETATION AND VALIDATION
	OF CLUSTER-ANALYSIS},
  journal = {Journal of Computational and Applied Mathematics},
  year = {1987},
  volume = {20},
  pages = {53-65},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.17}
}

@INPROCEEDINGS{Dudoit2001,
  author = {Dudoit, S. and Fridlyand, J.},
  title = {Bagging to improve the accuracy of a clustering procedure},
  booktitle = {NIPS Workshop on Machine Learning Techniques for Bioinformatics},
  year = {2001},
  pages = {1090-1099},
  address = {Whistler, Canada},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.17}
}

@ARTICLE{Gower1971,
  author = {Gower, J. C.},
  title = {GENERAL COEFFICIENT OF SIMILARITY AND SOME OF ITS PROPERTIES},
  journal = {Biometrics},
  year = {1971},
  volume = {27},
  pages = {857-\&},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.17}
}

@ARTICLE{Gower1986,
  author = {Gower, J. C. and Legendre, P.},
  title = {METRIC AND EUCLIDEAN PROPERTIES OF DISSIMILARITY COEFFICIENTS},
  journal = {Journal of Classification},
  year = {1986},
  volume = {3},
  pages = {5-48},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.17}
}

@INPROCEEDINGS{BenHur2002,
  author = {Ben-Hur},
  title = {A stability based method for discovering structure in clustered dataset},
  year = {2002},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.17}
}

@TECHREPORT{Fridlyand2001,
  author = {Fridlyand, J. and Dudoit, S.},
  title = {Applications of resampling methods to estimate the number of clusters
	and to improve the accuracy of a clustering method},
  year = {2001},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.17}
}

@ARTICLE{Breckenridge1989,
  author = {Breckenridge, J. N.},
  title = {REPLICATING CLUSTER-ANALYSIS - METHOD, CONSISTENCY, AND VALIDITY},
  journal = {Multivariate Behavioral Research},
  year = {1989},
  volume = {24},
  pages = {147-161},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.17}
}

@ARTICLE{Breckenridge2000,
  author = {Breckenridge, J. N.},
  title = {Validating cluster analysis: Consistent replication and symmetry},
  journal = {Multivariate Behavioral Research},
  year = {2000},
  volume = {35},
  pages = {261-285},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.17}
}

@ARTICLE{Bittner2000,
  author = {Bittner, M. and Meitzer, P. and Chen, Y. and Jiang, Y. and Seftor,
	E. and Hendrix, M. and Radmacher, M. and Simon, R. and Yakhini, Z.
	and Ben-Dor, A. and Sampas, N. and Dougherty, E. and Wang, E. and
	Marincola, F. and Gooden, C. and Lueders, J. and Glatfelter, A. and
	Pollock, P. and Carpten, J. and Gillanders, E. and Leja, D. and Dietrich,
	K. and Beaudry, C. and Berens, M. and Alberts, D. and Sondak, V.
	and Hayward, N. and Trent, J.},
  title = {Molecular classification of cutaneous malignant melanoma by gene
	expression profiling},
  journal = {Nature},
  year = {2000},
  volume = {406},
  pages = {536-540},
  number = {6795},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.17}
}

@ARTICLE{Kerr2001,
  author = {Kerr, M. K. and Churchill, G. A.},
  title = {Bootstrapping cluster analysis: Assessing the reliability of conclusions
	from microarray experiments},
  journal = {Proceedings of the National Academy of Sciences of the United States
	of America},
  year = {2001},
  volume = {98},
  pages = {8961-8965},
  number = {16},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.17}
}

@ARTICLE{Levine2001,
  author = {Levine, E. and Domany, E.},
  title = {Resampling method for unsupervised estimation of cluster validity},
  journal = {Neural Computation},
  year = {2001},
  volume = {13},
  pages = {2573-2593},
  number = {11},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.17}
}

@ARTICLE{Kell2004,
  author = {Kell, D. B. and Oliver, S. G.},
  title = {Here is the evidence, now what is the hypothesis? The complementary
	roles of inductive and hypothesis-driven science in the post-genomic
	era},
  journal = {Bioessays},
  year = {2004},
  volume = {26},
  pages = {99-105},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.17}
}

@ARTICLE{Lange2004,
  author = {Lange, T. and Roth, V. and Braun, M. L. and Buhmann, J. M.},
  title = {Stability-based validation of clustering solutions},
  journal = {Neural Computation},
  year = {2004},
  volume = {16},
  pages = {1299-1323},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.17}
}

@ELECTRONIC{UCIdatabase,
  author = {Hettich, S. and Bay, S. D.},
  year = {1999},
  title = {The UCI KDD Archive},
  organization = {Irvine, CA: University of California, Department of Information and
	Computer Science.},
  url = {http://kdd.ics.uci.edu/},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.20}
}

@ELECTRONIC{UCI-ML,
  author = {A. Asuncion and D.J. Newman},
  year = {2007},
  title = {{UCI} Machine Learning Repository},
  organization = {University of California, School of Information and Computer Science},
  url = {http://www.ics.uci.edu/$\sim$mlearn/{MLR}epository.html},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.20}
}

@ARTICLE{Hodson1970,
  author = {Hodson, F. R.},
  title = {CLUSTER ANALYSIS AND ARCHAEOLOGY - SOME NEW DEVELOPMENTS AND APPLICATIONS},
  journal = {World Archaeology},
  year = {1970},
  volume = {1},
  pages = {299-320},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.28}
}

@ARTICLE{Blashfield1979,
  author = {Blashfield, R. K. and Morey, L. C.},
  title = {CLASSIFICATION OF DEPRESSION THROUGH CLUSTER-ANALYSIS},
  journal = {Comprehensive Psychiatry},
  year = {1979},
  volume = {20},
  pages = {516-527},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.28}
}

@ARTICLE{Everitt2004,
  author = {Everitt, B.},
  title = {Untitled},
  journal = {Statistical Methods in Medical Research},
  year = {2004},
  volume = {13},
  pages = {343-345},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.28}
}

@ARTICLE{Gallegos2005,
  author = {Gallegos, M. T. and Ritter, G.},
  title = {A robust method for cluster analysis},
  journal = {Annals of Statistics},
  year = {2005},
  volume = {33},
  pages = {347-380},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.28}
}

@INPROCEEDINGS{Shim2005,
  author = {Shim, Y. and Chung, J. and Choi, I. C.},
  title = {A comparison study of cluster validity indices using a nonhierarchical
	clustering algorithm},
  booktitle = {International Conference on Computational Intelligence for Modelling,
	Control and Automation/International Conference on Intelligent Agents
	Web Technologies and International Commerce},
  year = {2005},
  editor = {Mohammadian, M.},
  pages = {199-203},
  address = {Vienna, AUSTRIA},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.28}
}

@ARTICLE{Mao1996,
  author = {Mao, J. C. and Jain, A. K.},
  title = {A self-organizing network for hyperellipsoidal clustering (HEC)},
  journal = {Ieee Transactions on Neural Networks},
  year = {1996},
  volume = {7},
  pages = {16-29},
  number = {1},
  abstract = {We propose a self-organizing network for hyperellipsoidal clustering
	(HEC), The HEC network consists of two layers, The first layer employs
	a number of principal component analysis subnetworks which are used
	to estimate the hyperellipsoidal shapes of currently formed clusters,
	The second layer then performs a competitive learning using the cluster
	shape information provided by the first layer. The HEC network performs
	a partitional clustering using the proposed regularized Mahalanobis
	distance. This regularized Mahalanobis distance is designed to deal
	with the problems in estimating the Mahalanobis distance when the
	number of patterns in a cluster is less than (ill-posed problem)
	or not considerably larger than (poorly posed problem) the dimensionality
	of the feature space during the clustering procedure. This regularized
	distance also achieves a tradeoff between hyperspherical and hyperellipsoidal
	cluster shapes so as to prevent the HEC network from producing usually
	large or unusually small clusters, The significance level of the
	Kolmogorov-Smirnov test on the distribution of the Mahalanobis distances
	of patterns in a cluster to the cluster center under the Gaussian
	cluster assumption is used as a compactness measure of the cluster,
	The HEC network has been tested on a number of artificial data sets
	and real data sets, We also apply the HEC network to texture segmentation
	problems. Experiments show that the HEC network leads to a significant
	improvement in the clustering results over the IC-means algorithm
	with Euclidean distance. Our results on real data sets also indicate
	that hyperellipsoidal shaped clusters are often encountered in practice.},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.28}
}

@ARTICLE{Mao1997,
  author = {Mao, J. C. and Jain, A.},
  title = {A self-organizing network for hyperellipsoidal clustering (HEC) -
	Reply},
  journal = {Ieee Transactions on Neural Networks},
  year = {1997},
  volume = {8},
  pages = {1563-1563},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.03.28}
}

@ARTICLE{Vlies2009,
  author = {van der Vlies, A. E. and Verwey, N. A. and Bouwman, F. H. and Blankenstein,
	M. A. and Klein, M. and Scheltens, P. and van der Flier, W. M.},
  title = {CSF biomarkers in relationship to cognitive profiles in Alzheimer
	disease},
  journal = {Neurology},
  year = {2009},
  volume = {72},
  pages = {1056-1061},
  number = {12},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.05}
}

@ARTICLE{Sellick1990,
  author = {Sellick, K. and Littlejohn, G. and Wallace, C. and Over, R.},
  title = {IDENTIFYING SUBCLASSES OF PATIENTS WITH RHEUMATOID-ARTHRITIS THROUGH
	CLUSTER-ANALYSIS},
  journal = {Journal of Rheumatology},
  year = {1990},
  volume = {17},
  pages = {1613-1619},
  number = {12},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.05}
}

@ARTICLE{Chaturvedi1997,
  author = {Chaturvedi, A. and Carroll, J. D. and Green, P. E. and Rotondo, J.
	A.},
  title = {A feature-based approach to market segmentation via overlapping K-centroids
	clustering},
  journal = {Journal of Marketing Research},
  year = {1997},
  volume = {34},
  pages = {370-377},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.05}
}

@ARTICLE{Koo2001,
  author = {Koo, E. and Nagy, Z. and Sesztak, M. and Ujfalussy, I. and Meretey,
	K. and Bohm, U. and Forgacs, S. and Szilagy, M. and Czirjak, L. and
	Farkas, V.},
  title = {Subsets in psoriatic arthritis formed by cluster analysis},
  journal = {Clinical Rheumatology},
  year = {2001},
  volume = {20},
  pages = {36-43},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.05}
}

@ARTICLE{Diallo-Danebrock2007,
  author = {Diallo-Danebrock, R. and Ting, E. and Gluz, O. and Herr, A. and Mohrmann,
	S. and Geddert, H. and Rody, A. and Schaefer, K. L. and Baldus, S.
	E. and Hartmann, A. and Wild, P. J. and Burson, M. and Gabbert, H.
	E. and Nitz, U. and Poremba, C.},
  title = {Protein expression profiling in high-risk breast cancer patients
	treated with high-dose or conventional dose-dense chemotherapy},
  journal = {Clinical Cancer Research},
  year = {2007},
  volume = {13},
  pages = {488-497},
  number = {2},
  note = {Part 1},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.05}
}

@ARTICLE{Brusco2008,
  author = {Brusco, M. J. and Kohn, H. F.},
  title = {Optimal partitioning of a data set based on the p-median model},
  journal = {Psychometrika},
  year = {2008},
  volume = {73},
  pages = {89-105},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.05}
}

@ARTICLE{Steinley2008b,
  author = {Steinley, D.},
  title = {Stability analysis in {$K$}-means clustering},
  journal = {British Journal of Mathematical \& Statistical Psychology},
  year = {2008},
  volume = {61},
  pages = {255-273},
  note = {Part 2},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.05}
}

@ARTICLE{Grun2009,
  author = {Grun, B. and Leisch, F.},
  title = {Dealing with label switching in mixture models under genuine multimodality},
  journal = {Journal of Multivariate Analysis},
  year = {2009},
  volume = {100},
  pages = {851-861},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.05}
}

@ARTICLE{Ubeyli,
  author = {\"{U}beyli, Elif and Do\u{g}du, Erdo\u{g}an},
  title = {Automatic Detection of Erythemato-Squamous Diseases Using k -Means
	Clustering},
  journal = {Journal of Medical Systems},
  year = {2009},
  note = {I can't find it in the list view of the journal (by volume, date),
	but can find by searching. Possibly is up early.},
  abstract = {A new approach based on the implementation of k-means clustering is
	presented for automated detection of erythemato-squamous diseases.
	The purpose of clustering techniques is to find a structure for the
	given data by finding similarities between data according to data
	characteristics. The studied domain contained records of patients
	with known diagnosis. The k-means clustering algorithm’s task was
	to classify the data points, in this case the patients with attribute
	data, to one of the five clusters. The algorithm was used to detect
	the five erythemato-squamous diseases when 33 features defining five
	disease indications were used. The purpose is to determine an optimum
	classification scheme for this problem. The present research demonstrated
	that the features well represent the erythemato-squamous diseases
	and the k-means clustering algorithm’s task achieved high classification
	accuracies for only five erythemato-squamous diseases.},
  citeulike-article-id = {3619137},
  doi = {10.1007/s10916-008-9229-6},
  owner = {rebeccaspeckman},
  posted-at = {2008-11-20 20:23:48},
  timestamp = {2009.04.09}
}

@ARTICLE{Goodman1954,
  author = {Goodman, L. A. and Kruskal, W. H.},
  title = {MEASURES OF ASSOCIATION FOR CROSS CLASSIFICATIONS},
  journal = {Journal of the American Statistical Association},
  year = {1954},
  volume = {49},
  pages = {732-764},
  number = {268},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Mirkin2001,
  author = {Mirkin, B.},
  title = {Eleven ways to look at the chi-squared coefficient for contingency
	tables},
  journal = {American Statistician},
  year = {2001},
  volume = {55},
  pages = {111-120},
  number = {2},
  abstract = {This article has been written in recognition of the 100th anniversary
	of introduction of the concept of association between categorical
	variables by Yule and Pearson. The most popular among the contingency
	coefficients, Pearson's chi-squared, estimates the bias of a cross-classification
	from the statistical independence. Also, it measures association
	per se between the row and column variables. The purpose of this
	article is to present a collection of ii definitions for the chi-squared
	coefficient related to either of these goals. One of the quoted definitions
	of the chi-squared coefficient seems especially appealing as an association
	measure: the averaged relative Quetelet index of category-to-category
	associations.},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Topchy2005,
  author = {Topchy, A. and Jain, A. K. and Punch, W.},
  title = {Clustering ensembles: Models of consensus and weak partitions},
  journal = {Ieee Transactions on Pattern Analysis and Machine Intelligence},
  year = {2005},
  volume = {27},
  pages = {1866-1881},
  number = {12},
  abstract = {Clustering ensembles have emerged as a powerful method for improving
	both the robustness as well as the stability of unsupervised classification
	solutions. However, finding a consensus clustering from multiple
	partitions is a difficult problem that can be approached from graph-based,
	combinatorial, or statistical perspectives. This study extends previous
	research on clustering ensembles in several respects. First, we introduce
	a unified representation for multiple clusterings and formulate the
	corresponding categorical clustering problem. Second, we propose
	a probabilistic model of consensus using a finite mixture of multinomial
	distributions in a space of clusterings. A combined partition is
	found as a solution to the corresponding maximum-likelihood problem
	using the EM algorithm. Third, we define a new consensus function
	that is related to the classical intraclass variance criterion using
	the generalized mutual information definition. Finally, we demonstrate
	the efficacy of combining partitions generated by weak clustering
	algorithms that use data projections and random data splits. A simple
	explanatory model is offered for the behavior of combinations of
	such weak clustering components. Combination accuracy is analyzed
	as a function of several parameters that control the power and resolution
	of component partitions as well as the number of partitions. We also
	analyze clustering ensembles with incomplete information and the
	effect of missing cluster labels on the quality of overall consensus.
	Experimental results demonstrate the effectiveness of the proposed
	methods on several real-world data sets.},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Albatineh2006,
  author = {Albatineh, A. N. and Niewiadomska-Bugaj, M. and Mihalko, D.},
  title = {On similarity indices and correction for chance agreement},
  journal = {Journal of Classification},
  year = {2006},
  volume = {23},
  pages = {301-313},
  number = {2},
  abstract = {Similarity indices can be used to compare partitions (clusterings)
	of a data set. Many such indices were introduced in the literature
	over the years. We are showing that out of 28 indices we were able
	to track, there are 22 different ones. Even though their values differ
	for the same clusterings compared, after correcting for agreement
	attributed to chance only, their values become similar and some of
	them even become equivalent. Consequently, the problem of choice
	of the index to be used for comparing different clusterings becomes
	less important.},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Warrens2008,
  author = {Warrens, M. J.},
  title = {On the indeterminacy of resemblance measures for binary (presence/absence)
	data},
  journal = {Journal of Classification},
  year = {2008},
  volume = {25},
  pages = {125-136},
  number = {1},
  abstract = {Many similarity coefficients for binary data are defined as fractions.
	For certain resemblance measures the denominator may become zero.
	If the denominator is zero the value of the coefficient is indeterminate.
	It is shown that the seriousness of the indeterminacy problem differs
	with the resemblance measures. Following Batagelj and Bren (1995)
	we remove the indeterminacies by defining appropriate values in critical
	cases.},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Warrens2008a,
  author = {Warrens, M. J.},
  title = {On similarity coefficients for 2 x 2 tables and correction for chance},
  journal = {Psychometrika},
  year = {2008},
  volume = {73},
  pages = {487-502},
  number = {3},
  abstract = {This paper studies correction for chance in coefficients that are
	linear functions of the observed proportion of agreement. The paper
	unifies and extends various results on correction for chance in the
	literature. A specific class of coefficients is used to illustrate
	the results derived in this paper. Coefficients in this class, e.g.
	the simple matching coefficient and the Dice/Sorenson coefficient,
	become equivalent after correction for chance, irrespective of what
	expectation is used. The coefficients become either Cohen's kappa,
	Scott's pi, Mak's rho, Goodman and Kruskal's lambda, or Hamann's
	eta, depending on what expectation is considered appropriate. Both
	a multicategorical generalization and a multivariate generalization
	are discussed.},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Warrens2008b,
  author = {Warrens, M.},
  title = {On Association Coefficients for 2x2 Tables and Properties That Do
	Not Depend on the Marginal Distributions},
  journal = {Psychometrika},
  year = {2008},
  volume = {73},
  pages = {777-789},
  number = {4},
  abstract = {We discuss properties that association coefficients may have in general,
	e.g., zero value under statistical independence, and we examine coefficients
	for 2x2 tables with respect to these properties. Furthermore, we
	study a family of coefficients that are linear transformations of
	the observed proportion of agreement given the marginal probabilities.
	This family includes the phi coefficient and Cohen's kappa. The main
	result is that the linear transformations that set the value under
	independence at zero and the maximum value at unity, transform all
	coefficients in this family into the same underlying coefficient.
	This coefficient happens to be Loevinger's H.},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Warrens2008c,
  author = {Warrens, M. J.},
  title = {Bounds of Resemblance Measures for Binary (Presence/Absence) Variables},
  journal = {Journal of Classification},
  year = {2008},
  volume = {25},
  pages = {195-208},
  number = {2},
  abstract = {Bounds of association coefficients for binary variables are derived
	using the arithmetic-geometric-harmonic mean inequality. More precisely,
	it is shown which presence/absence coefficients are bounds with respect
	to each other. Using the new bounds it is investigated whether a
	coefficient is in general closer to either its upper or its lower
	bound.},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Wu2009,
  author = {Wu, J. J. and Chen, J. and Xiong, H. and Xie, M.},
  title = {External validation measures for K-means clustering: A data distribution
	perspective},
  journal = {Expert Systems with Applications},
  year = {2009},
  volume = {36},
  pages = {6050-6061},
  number = {3},
  note = {Part 2},
  abstract = {Cluster validation is an important part of any cluster analysis. External
	measures such as entropy, purity and mutual information are often
	used to evaluate K-means clustering. However, whether these measures
	are indeed suitable for K-means clustering remains unknown. Along
	this line, in this paper, we show that a data distribution view is
	of great use to selecting the right measures for K-means clustering.
	Specifically, we first introduce the data distribution view of K-means,
	and the resultant uniform effect on highly imbalanced data sets.
	Eight external measures widely used in recent data mining tasks are
	also collected as candidates for K-means evaluation. Then, we demonstrate
	that only three measures, namely the variation of information (VI),
	the van Dongen criterion (VD) and the Mirkin metric (M), can detect
	the negative uniform effect of K-means in the clustering results.
	We also provide new normalization schemes for these three measures,
	i.e., VI'(norm), VD'(norm) and M'(norm), which enables the cross-data
	comparisons of clustering qualities. Finally, we explore some properties
	such as the consistency and sensitivity of the three measures, and
	give some advice on how to use them in K-means practice. (C) 2008
	Elsevier Ltd. All rights reserved.},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Makuch1989,
  author = {Makuch, R. W. and Rosenberg, P. S. and Scott, G.},
  title = {GOODMAN AND KRUSKAL LAMBDA - A NEW LOOK AT AN OLD MEASURE OF ASSOCIATION},
  journal = {Statistics in Medicine},
  year = {1989},
  volume = {8},
  pages = {619-631},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Goodman1959,
  author = {Goodman, L. A. and Kruskal, W. H.},
  title = {MEASURES OF ASSOCIATION FOR CROSS CLASSIFICATIONS .2. FURTHER DISCUSSION
	AND REFERENCES},
  journal = {Journal of the American Statistical Association},
  year = {1959},
  volume = {54},
  pages = {123-163},
  number = {285},
  note = {Times Cited: 203},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Goodman1972,
  author = {Goodman, L. A. and Kruskal, W. H.},
  title = {MEASURES OF ASSOCIATION FOR CROSS CLASSIFICATIONS .4. SIMPLIFICATION
	OF ASYMPTOTIC VARIANCES},
  journal = {Journal of the American Statistical Association},
  year = {1972},
  volume = {67},
  pages = {415-\&},
  number = {33},
  note = {Times Cited: 148},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Gray1975,
  author = {Gray, L. N. and Campbell, R.},
  title = {STATISTICAL SIGNIFICANCE OF LAMBDA COEFFICIENTS - A COMMENT},
  journal = {Behavioral Science},
  year = {1975},
  volume = {20},
  pages = {258-259},
  number = {4},
  note = {Times Cited: 1},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Goodman1963,
  author = {Goodman, L. A. and Kruskal, W. H.},
  title = {MEASURES OF ASSOCIATION FOR CROSS CLASSIFICATIONS .3. APPROXIMATE
	SAMPLING THEORY},
  journal = {Journal of the American Statistical Association},
  year = {1963},
  volume = {58},
  pages = {310-\&},
  number = {302},
  note = {Times Cited: 273},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Meila2001,
  author = {Meila, M. and Heckerman, D.},
  title = {An experimental comparison of model-based clustering methods},
  journal = {Machine Learning},
  year = {2001},
  volume = {42},
  pages = {9-29},
  number = {1-2},
  note = {Times Cited: 37},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Meila2003,
  author = {Meila, M.},
  title = {Comparing clusterings by the variation of information},
  journal = {Learning Theory and Kernel Machines},
  year = {2003},
  volume = {2777},
  pages = {173-187},
  note = {Times Cited: 7 Scholkopf, B Warmuth, MK 16th Annual Conference on
	Learning Theory/7th Annual Workshop on Kernel Machines AUG 24-27,
	2003 WASHINGTON, D.C.},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Patrikainen2006,
  author = {Patrikainen, A. and Meila, M.},
  title = {Comparing subspace clusterings},
  journal = {Ieee Transactions on Knowledge and Data Engineering},
  year = {2006},
  volume = {18},
  pages = {902-916},
  number = {7},
  note = {Times Cited: 5},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Wu2007,
  author = {Wu, J. J. and Xiong, H. and Chen, J. and Zhou, W. J.},
  title = {A generalization of proximity functions for K-means},
  journal = {Icdm 2007: Proceedings of the Seventh Ieee International Conference
	on Data Mining},
  year = {2007},
  pages = {361-370},
  note = {Times Cited: 0 Ramakrishnan, N Zaiane, OR Shi, Y Clifton, CW Wu,
	XD 7th IEEE International Conference on Data Mining OCT 28-31, 2007
	Omaha, NE},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Meila2007,
  author = {Meila, M.},
  title = {Comparing clusterings - an information based distance},
  journal = {Journal of Multivariate Analysis},
  year = {2007},
  volume = {98},
  pages = {873-895},
  number = {5},
  note = {Times Cited: 11},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Suich2003,
  author = {Suich, R. C. and Turek, R. J.},
  title = {An asymptotic partial correlation test for the Goodman-Kruskal lambda},
  journal = {British Journal of Mathematical \& Statistical Psychology},
  year = {2003},
  volume = {56},
  pages = {111-117},
  note = {Part 1},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Edwardes2000,
  author = {Edwardes, M. D. D. and Baltzan, M.},
  title = {The generalization of the odds ratio, risk ratio and risk difference
	to r x k tables},
  journal = {Statistics in Medicine},
  year = {2000},
  volume = {19},
  pages = {1901-1914},
  number = {14},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Sung1998,
  author = {Sung, C. S. and Ahn, S. J.},
  title = {A proportional-reduction-in-impurity measure of association for categorical
	variables},
  journal = {Communications in Statistics-Theory and Methods},
  year = {1998},
  volume = {27},
  pages = {2083-2110},
  number = {8},
  note = {Times Cited: 0},
  abstract = {This paper presents a proportional-reduction-in-impurity (PRI) measure
	for categorical association, that employs application-dependent loss
	functions which make the measure widely applicable. The well-known
	proportional-reduction-in-error (PRE) measure is shown to be a special
	case of the new PRI measure. Moreover, the asymptotic variance of
	the maximum likelihood estimator (MLE) of the measure is derived
	to facilitate its use for statistical inference. An extension of
	the PRI measure to compositional association is made to show that
	it can have a variety of applications. Selected loss functions are
	treated to illustrate the derivation of the measure.},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.13}
}

@ARTICLE{Higham2002,
  author = {Higham, N. J.},
  title = {Computing the nearest correlation matrix - a problem from finance},
  journal = {Ima Journal of Numerical Analysis},
  year = {2002},
  volume = {22},
  pages = {329-343},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.14}
}

@ARTICLE{Grubisic2007,
  author = {Grubisic, I. and Pietersz, R.},
  title = {Efficient rank reduction of correlation matrices},
  journal = {Linear Algebra and Its Applications},
  year = {2007},
  volume = {422},
  pages = {629-653},
  number = {2-3},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.14}
}

@ARTICLE{Gerbing1987,
  author = {Gerbing, D. W. and Anderson, J. C.},
  title = {IMPROPER SOLUTIONS IN THE ANALYSIS OF COVARIANCE-STRUCTURES - THEIR
	INTERPRETABILITY AND A COMPARISON OF ALTERNATE RESPECIFICATIONS},
  journal = {Psychometrika},
  year = {1987},
  volume = {52},
  pages = {99-111},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.14}
}

@ARTICLE{Anderson1988,
  author = {Anderson, J. C. and Gerbing, D. W.},
  title = {STRUCTURAL EQUATION MODELING IN PRACTICE - A REVIEW AND RECOMMENDED
	2-STEP APPROACH},
  journal = {Psychological Bulletin},
  year = {1988},
  volume = {103},
  pages = {411-423},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.14}
}

@INBOOK{Jollois,
  chapter = {Assessing the number of clusters in the latent class model},
  title = {Book},
  author = {Jollois, Nadif, and Govaert},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.14}
}

@INPROCEEDINGS{Gan2005,
  author = {Gan, G. J. and Yang, Z. J. and Wu, J. H.},
  title = {A genetic k-modes algorithm for clustering categorical data},
  booktitle = {1st International Conference on Advanced Data Mining and Applications},
  year = {2005},
  editor = {Li, X. and Wang, S. and Dong, Z. Y.},
  pages = {195-202},
  address = {Wuhan, PEOPLES R CHINA},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.15}
}

@INCOLLECTION{Luo2006,
  author = {Luo, H. L. and Kong, F. S. and Li, Y. X.},
  title = {Combining multiple clusterings via k-modes algorithm},
  booktitle = {Advanced Data Mining and Applications, Proceedings},
  year = {2006},
  volume = {4093},
  series = {Lecture Notes in Artificial Intelligence},
  pages = {308-315},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.15}
}

@BOOK{Bollen1993,
  title = {Testing structural equation models},
  publisher = {Sage Publications},
  year = {1993},
  author = {Bollen, Kenneth A. and Long, J. Scott},
  address = {Newbury Park},
  note = {92037324 Kenneth A. Bollen, J. Scott Long, editors. ill. ; 23 cm.
	Sage focus editions ; v. 154 Includes bibliographical references.},
  keywords = {Social sciences Mathematical models. Social sciences Methodology.},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.15}
}

@BOOK{Jain1988,
  title = {Algorithms for clustering data},
  publisher = {Prentice Hall},
  year = {1988},
  author = {Jain, Anil K. and Dubes, Richard C.},
  address = {Englewood Cliffs, N.J.},
  note = {87024500 Anil K. Jain, Richard C. Dubes. ill. ; 24 cm. Prentice Hall
	advanced reference series Bibliography: p. 275-296. Includes indexes.},
  keywords = {Cluster analysis Data processing. Computer algorithms.},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.15}
}

@INBOOK{Wothke1993,
  chapter = {Nonpositive definite matrices in structural modeling},
  pages = {256-293},
  title = {Testing structural equation models},
  publisher = {Sage Publications},
  year = {1993},
  editor = {Bollen, Kenneth A. and Long, J. Scott},
  author = {Wothke, Werner},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.15}
}

@ARTICLE{Joreskog1967,
  author = {Joreskog, K. G.},
  title = {SOME CONTRIBUTIONS TO MAXIMUM LIKELIHOOD FACTOR ANALYSIS},
  journal = {Psychometrika},
  year = {1967},
  volume = {32},
  pages = {443-\&},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.15}
}

@ARTICLE{Lawley1973,
  author = {Lawley, D. N. and Maxwell, A. E.},
  title = {REGRESSION AND FACTOR-ANALYSIS},
  journal = {Biometrika},
  year = {1973},
  volume = {60},
  pages = {331-338},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.15}
}

@ARTICLE{Qi2006,
  author = {Qi, H. D. and Sun, D. F.},
  title = {A quadratically convergent Newton method for computing the nearest
	correlation matrix},
  journal = {Siam Journal on Matrix Analysis and Applications},
  year = {2006},
  volume = {28},
  pages = {360-385},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.15}
}

@INPROCEEDINGS{Munoz2006,
  author = {Munoz, A. and de Diego, I. M.},
  title = {From indefinite to positive semi-definite matrices},
  booktitle = {Joint International Workshop on Structural, Syntactic, and Statistical
	Pattern Recognition},
  year = {2006},
  editor = {Yeung, D. Y. and Kwok, J. T. and Fred, A. and Roli, F. and DeRidder,
	D.},
  pages = {764-772},
  address = {Hong Kong, PEOPLES R CHINA},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.15}
}

@ARTICLE{Qi2007,
  author = {Qi, H. D. and Xia, Z. G. and Xing, G. M.},
  title = {An application of the nearest correlation matrix on web document
	classification},
  journal = {Journal of Industrial and Management Optimization},
  year = {2007},
  volume = {3},
  pages = {701-713},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.04.15}
}

@UNPUBLISHED{Yeung2001b,
  author = {Ka Yee Yeung and Walter L. Ruzzo},
  title = {Details of the adjusted Rand index and clustering algorithms},
  year = {2001},
  issue = {0},
  journal = {Unpublished supplement to Bioinformatics paper},
  pages = {0},
  volume = {0}
}

@ARTICLE{Hubert1977,
  author = {Lawrence Hubert},
  title = {Nominal scale response agreement as a generalized correlation},
  journal = {British Journal of Mathematical and Statistical Psychology},
  year = {1977},
  volume = {30},
  pages = {98--103}
}

@ARTICLE{Hubert1979,
  author = {Lawrence Hubert},
  title = {Matching models in the analysis of cross-classifications},
  journal = {Psychometrika},
  year = {1979},
  volume = {44},
  pages = {21--41},
  issue = {1}
}

@ARTICLE{Tan1982,
  author = {Tan, E. M. and Cohen, A. S. and Fries, J. F. and Masi, A. T. and
	McShane, D. J. and Rothfield, N. F. and Schaller, J. G. and Talal,
	N. and Winchester, R. J.},
  title = {The 1982 revised criteria for the classification of systemic lupus
	erythematosus},
  journal = {Arthritis Rheum},
  year = {1982},
  volume = {25},
  pages = {1271-7},
  number = {11},
  abstract = {The 1971 preliminary criteria for the classification of systemic lupus
	erythematosus (SLE) were revised and updated to incorporate new immunologic
	knowledge and improve disease classification. The 1982 revised criteria
	include fluorescence antinuclear antibody and antibody to native
	DNA and Sm antigen. Some criteria involving the same organ systems
	were aggregated into single criteria. Raynaud's phenomenon and alopecia
	were not included in the 1982 revised criteria because of low sensitivity
	and specificity. The new criteria were 96% sensitive and 96% specific
	when tested with SLE and control patient data gathered from 18 participating
	clinics. When compared with the 1971 criteria, the 1982 revised criteria
	showed gains in sensitivity and specificity.},
  keywords = {Arthritis/etiology Diagnosis, Differential False Positive Reactions
	Hematologic Diseases/etiology Humans Kidney Diseases/etiology Lupus
	Erythematosus, Systemic/*classification/diagnosis Mouth Diseases/etiology
	Nervous System Diseases/etiology Serologic Tests Serositis/etiology
	Skin Diseases/etiology Ulcer/etiology}
}

@BOOK{Fletcher1988,
  title = {Clinical epidemiology : the essentials},
  publisher = {Williams \& Wilkins},
  year = {1988},
  author = {Fletcher, Robert H. and Fletcher, Suzanne W. and Wagner, Edward H.},
  address = {Baltimore},
  edition = {2nd},
  note = {87010445 Robert H. Fletcher, Suzanne W. Fletcher, Edward H. Wagner.
	ill. ; 23 cm. Includes bibliographies and index.},
  keywords = {Clinical epidemiology. Epidemiologic Methods.},
  owner = {rebeccaspeckman},
  timestamp = {2009.05.25}
}

@BOOK{Weiss2002,
  title = {Introductory statistics},
  publisher = {Addison-Wesley},
  year = {2002},
  author = {Weiss, N. A.},
  address = {Boston},
  edition = {6th},
  note = {2001022689 Neil A. Weiss ; biographies by Carol A. Weiss. ill. (some
	col.) ; 26 cm. + 1 computer optical disc (4 3/4 in.) Includes indexes.},
  keywords = {Statistics.},
  owner = {rebeccaspeckman},
  timestamp = {2009.05.25}
}

@ARTICLE{Speckman2009a,
  author = {Speckman, R. A.},
  title = {A comparison of $K$-centroids variations for the identification of
	disease subtypes with presence/absence attributes.},
  journal = {in preparation},
  year = {2009},
  owner = {rebeccaspeckman},
  timestamp = {2009.05.28}
}

@ARTICLE{Akaike1974,
  author = {Akaike, H.},
  title = {NEW LOOK AT STATISTICAL-MODEL IDENTIFICATION},
  journal = {Ieee Transactions on Automatic Control},
  year = {1974},
  volume = {AC19},
  pages = {716-723},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.08}
}

@ARTICLE{Schwarz1978,
  author = {Schwarz, G.},
  title = {ESTIMATING DIMENSION OF A MODEL},
  journal = {Annals of Statistics},
  year = {1978},
  volume = {6},
  pages = {461-464},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.08}
}

@ARTICLE{Kass1995,
  author = {Kass, R. E. and Raftery, A. E.},
  title = {BAYES FACTORS},
  journal = {Journal of the American Statistical Association},
  year = {1995},
  volume = {90},
  pages = {773-795},
  number = {430},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.08}
}

@ARTICLE{Tibshirani1997,
  author = {Tibshirani, R.},
  title = {The lasso method for variable selection in the cox model},
  journal = {Statistics in Medicine},
  year = {1997},
  volume = {16},
  pages = {385-395},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.08}
}

@ARTICLE{Volinsky1997,
  author = {Volinsky, C. T. and Madigan, D. and Raftery, A. E. and Kronmal, R.
	A.},
  title = {Bayesian model averaging in proportional hazard models. Assessing
	the risk of a stroke},
  journal = {Applied Statistics-Journal of the Royal Statistical Society Series
	C},
  year = {1997},
  volume = {46},
  pages = {433-448},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.08}
}

@ARTICLE{Volinsky2000,
  author = {Volinsky, C. T. and Raftery, A. E.},
  title = {Bayesian information criterion for censored survival models},
  journal = {Biometrics},
  year = {2000},
  volume = {56},
  pages = {256-262},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.08}
}

@ARTICLE{Ibrahim1999,
  author = {Ibrahim, J. G. and Chen, M. H. and MacEachern, S. N.},
  title = {Bayesian variable selection for proportional hazards models},
  journal = {Canadian Journal of Statistics-Revue Canadienne De Statistique},
  year = {1999},
  volume = {27},
  pages = {701-717},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.08}
}

@ARTICLE{Fan2002,
  author = {Fan, J. Q. and Li, R. Z.},
  title = {Variable selection for Cox's proportional hazards model and frailty
	model},
  journal = {Annals of Statistics},
  year = {2002},
  volume = {30},
  pages = {74-99},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.08}
}

@ARTICLE{Box-Steffensmeier2006,
  author = {Box-Steffensmeier, J. M. and De Boef, S.},
  title = {Repeated events survival models: The conditional frailty model},
  journal = {Statistics in Medicine},
  year = {2006},
  volume = {25},
  pages = {3518-3533},
  number = {20},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.08}
}

@ARTICLE{Liang2008,
  author = {Liang, H. and Zou, G.},
  title = {Improved AIC selection strategy for survival analysis},
  journal = {Computational Statistics \& Data Analysis},
  year = {2008},
  volume = {52},
  pages = {2538-2548},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.08}
}

@ARTICLE{Lin1994,
  author = {Lin, D. Y.},
  title = {COX REGRESSION-ANALYSIS OF MULTIVARIATE FAILURE TIME DATA - THE MARGINAL
	APPROACH},
  journal = {Statistics in Medicine},
  year = {1994},
  volume = {13},
  pages = {2233-2247},
  number = {21},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.10}
}

@ARTICLE{Pinner1996,
  author = {Pinner, R. W. and Teutsch, S. M. and Simonsen, L. and Klug, L. A.
	and Graber, J. M. and Clarke, M. J. and Berkelman, R. L.},
  title = {Trends in infectious diseases mortality in the United States},
  journal = {Jama-Journal of the American Medical Association},
  year = {1996},
  volume = {275},
  pages = {189-193},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.10}
}

@ARTICLE{Wilson1998,
  author = {Wilson, D. and Bhopal, R.},
  title = {Impact of infection on mortality and hospitalization in the North
	East of England},
  journal = {Journal of Public Health Medicine},
  year = {1998},
  volume = {20},
  pages = {386-395},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.10}
}

@ARTICLE{Lin2000,
  author = {Lin, D. Y. and Wei, L. J. and Yang, I. and Ying, Z.},
  title = {Semiparametric regression for the mean and rate functions of recurrent
	events},
  journal = {Journal of the Royal Statistical Society Series B-Statistical Methodology},
  year = {2000},
  volume = {62},
  pages = {711-730},
  note = {Part 4},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.10}
}

@ARTICLE{Lu2005,
  author = {Lu, T. H.},
  title = {Don't overlook infectious diseases in ICD-9 chapters other than Chapter
	I},
  journal = {International Journal of Infectious Diseases},
  year = {2005},
  volume = {9},
  pages = {180-181},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.10}
}

@ARTICLE{Turner1929,
  author = {Turner, G. G.},
  title = {Cancer of the colon - Being the Annual oration of the Medical Society
	of London, delivered on May 13th, 1929},
  journal = {Lancet},
  year = {1929},
  volume = {1},
  pages = {1017-1023},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.13}
}

@ARTICLE{Bleicher1949,
  author = {Bleicher, J. E.},
  title = {CANCER OF THE COLON AND RECTUM - A 12-YEAR SURVEY OF 142 CASES IN
	A GENERAL HOSPITAL},
  journal = {Cancer},
  year = {1949},
  volume = {2},
  pages = {25-27},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.13}
}

@ARTICLE{Bader1952,
  author = {Bader, G. M. and Papanicolaou, G. N.},
  title = {THE APPLICATION OF CYTOLOGY IN THE DIAGNOSIS OF CANCER OF THE RECTUM,
	SIGMOID, AND DESCENDING COLON},
  journal = {Cancer},
  year = {1952},
  volume = {5},
  pages = {307-314},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.13}
}

@ARTICLE{Spratt1958,
  author = {Spratt, J. S. and Ackerman, L. V. and Moyer, C. A.},
  title = {RELATIONSHIP OF POLYPS OF THE COLON TO COLONIC CANCER},
  journal = {Annals of Surgery},
  year = {1958},
  volume = {148},
  pages = {682-698},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.13}
}

@ARTICLE{Floyd1966,
  author = {Floyd, C. E. and Stirling, C. T. and Cohn, I.},
  title = {CANCER OF COLON RECTUM AND ANUS - REVIEW OF 1,687 CASES},
  journal = {Annals of Surgery},
  year = {1966},
  volume = {163},
  pages = {829-\&},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.13}
}

@ARTICLE{Cole1966,
  author = {Cole, W. H. and Roberts, S. S. and Strehl, F. W.},
  title = {MODERN CONCEPTS IN CANCER OF COLON AND RECTUM},
  journal = {Cancer},
  year = {1966},
  volume = {19},
  pages = {1347-\&},
  number = {10},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.13}
}

@ARTICLE{Wynder1967,
  author = {Wynder, E. L. and Shigemat.T},
  title = {ENVIRONMENTAL FACTORS OF CANCER OF COLON AND RECTUM},
  journal = {Cancer},
  year = {1967},
  volume = {20},
  pages = {1520-\&},
  number = {9},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.13}
}

@ARTICLE{Wynder1969,
  author = {Wynder, E. L. and Kajitani, T. and Ishikawa, S. and Dodo, H. and
	Takano, A.},
  title = {ENVIRONMENTAL FACTORS OF CANCER OF COLON AND RECTUM .2. JAPANESE
	EPIDEMIOLOGICAL DATA},
  journal = {Cancer},
  year = {1969},
  volume = {23},
  pages = {1210-\&},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.13}
}

@ARTICLE{Lynch1973,
  author = {Lynch, H. T. and Guirgis, H. and Swartz, M. and Lynch, J. and Krush,
	A. J. and Kaplan, A. R.},
  title = {GENETICS AND COLON CANCER},
  journal = {Archives of Surgery},
  year = {1973},
  volume = {106},
  pages = {669-675},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.13}
}

@ARTICLE{Wolff1974,
  author = {Wolff, W. I. and Shinya, H.},
  title = {EARLIER DIAGNOSIS OF CANCER OF COLON THROUGH COLONIC ENDOSCOPY (COLONOSCOPY)},
  journal = {Cancer},
  year = {1974},
  volume = {34},
  pages = {912-931},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.13}
}

@ARTICLE{Burkitt1971,
  author = {Burkitt, D. P.},
  title = {EPIDEMIOLOGY OF CANCER OF COLON AND RECTUM},
  journal = {Cancer},
  year = {1971},
  volume = {28},
  pages = {3-\&},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.13}
}

@ARTICLE{Morson1974,
  author = {Morson, B. C.},
  title = {EVOLUTION OF CANCER OF COLON AND RECTUM},
  journal = {Cancer},
  year = {1974},
  volume = {34},
  pages = {845-849},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.13}
}

@ARTICLE{Brenner1990,
  author = {Brenner, H. and Savitz, D. A.},
  title = {THE EFFECTS OF SENSITIVITY AND SPECIFICITY OF CASE SELECTION ON VALIDITY,
	SAMPLE-SIZE, PRECISION, AND POWER IN HOSPITAL-BASED CASE-CONTROL
	STUDIES},
  journal = {American Journal of Epidemiology},
  year = {1990},
  volume = {132},
  pages = {181-192},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.13}
}

@ARTICLE{Lasky1994,
  author = {Lasky, T. and Stolley, P. D.},
  title = {SELECTION OF CASES AND CONTROLS},
  journal = {Epidemiologic Reviews},
  year = {1994},
  volume = {16},
  pages = {6-17},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.13}
}

@ARTICLE{Everitt1995,
  author = {Everitt, B. S.},
  title = {CLASSIFICATION AND CLUSTER-ANALYSIS - COMMENTARY},
  journal = {British Medical Journal},
  year = {1995},
  volume = {311},
  pages = {535-536},
  number = {7004},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Dollfus1996,
  author = {Dollfus, S. and Everitt, B. and Ribeyre, J. M. and AssoulyBesse,
	F. and Sharp, C. and Petit, M.},
  title = {Identifying subtypes of schizophrenia by cluster analyses},
  journal = {Schizophrenia Bulletin},
  year = {1996},
  volume = {22},
  pages = {545-555},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Jones2002,
  author = {Jones, E. and Hodgins-Vermaas, R. and McCartney, H. and Everitt,
	B. and Beech, C. and Poynter, D. and Palmer, I. and Hyams, K. and
	Wessely, S.},
  title = {Post-combat syndromes from the Boer war to the Gulf war: A cluster
	analysis of their nature and attribution},
  journal = {British Medical Journal},
  year = {2002},
  volume = {324},
  pages = {321-324},
  number = {7333},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Everitt2002,
  author = {Everitt, B. and Ismail, K. and David, A. S. and Wessely, S.},
  title = {Searching for a Gulf War syndrome using cluster analysis},
  journal = {Psychological Medicine},
  year = {2002},
  volume = {32},
  pages = {1371-1378},
  number = {8},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Everitt1971,
  author = {Everitt, B. S. and Gourlay, A. J. and Kendell, R. E.},
  title = {ATTEMPT AT VALIDATION OF TRADITIONAL PSYCHIATRIC SYNDROMES BY CLUSTER
	ANALYSIS},
  journal = {British Journal of Psychiatry},
  year = {1971},
  volume = {119},
  pages = {399-\&},
  number = {551},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Arnaut-Rollier1999,
  author = {Arnaut-Rollier, I. and Vauterin, L. and De Vos, P. and Massart, D.
	L. and Devriese, L. A. and De Zutter, L. and Van Hoof, J.},
  title = {A numerical taxonomic study of the Pseudomonas flora isolated from
	poultry meat},
  journal = {Journal of Applied Microbiology},
  year = {1999},
  volume = {87},
  pages = {15-28},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Iruela2002,
  author = {Iruela, M. and Rubio, J. and Cubero, J. I. and Gil, J. and Millan,
	T.},
  title = {Phylogenetic analysis in the genus {C}icer and cultivated chickpea
	using {RAPD} and {ISSR} markers},
  journal = {Theoretical and Applied Genetics},
  year = {2002},
  volume = {104},
  pages = {643-651},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@INPROCEEDINGS{Bayati2008,
  author = {Bayati, H. and Davoudi, H. and Fatemizadeh, E.},
  title = {A Heuristic Method for Finding the Optimal Number of Clusters with
	Application in Medical Data},
  booktitle = {30th Annual International Conference of the IEEE-Engineering-in-Medicine-and-Biology-Society},
  year = {2008},
  pages = {4684-4687},
  address = {Vancouver, CANADA},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Souto2008,
  author = {de Souto, M. C. P. and Costa, I. G. and de Araujo, D. S. A. and Ludermir,
	T. B. and Schliep, A.},
  title = {Clustering cancer gene expression data: a comparative study},
  journal = {Bmc Bioinformatics},
  year = {2008},
  volume = {9},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Ibanez2009,
  author = {Ibanez, M. T. and Caru, M. and Herrera, M. and Gonzalez, L. and Martin,
	L. and Miranda, J. and Navarro-Cerrillo, R.},
  title = {Clones identification of Sequoia sempervirens (D. Don) Endl. in Chile
	by using PCR-RAPDs technique},
  journal = {Journal of Zhejiang University-Science B},
  year = {2009},
  volume = {10},
  pages = {112-119},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Kurihara2009,
  author = {Kurihara, K. and Welling, M.},
  title = {Bayesian k-Means as a "Maximization-Expectation" Algorithm},
  journal = {Neural Computation},
  year = {2009},
  volume = {21},
  pages = {1145-1172},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Piot1980,
  author = {Piot, P. and Vandyck, E. and Goodfellow, M. and Falkow, S.},
  title = {A TAXONOMIC STUDY OF GARDNERELLA-VAGINALIS (HEMOPHILUS, VAGINALIS)
	GARDNER AND DUKES 1955},
  journal = {Journal of General Microbiology},
  year = {1980},
  volume = {119},
  pages = {373-\&},
  number = {AUG},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Martin1981,
  author = {Martin, R. and Riley, P. S. and Hollis, D. G. and Weaver, R. E. and
	Krichevsky, M. I.},
  title = {CHARACTERIZATION OF SOME GROUPS OF GRAM-NEGATIVE NONFERMENTATIVE
	BACTERIA BY THE CARBON SOURCE ALKALINIZATION TECHNIQUE},
  journal = {Journal of Clinical Microbiology},
  year = {1981},
  volume = {14},
  pages = {39-47},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Goodfellow1982,
  author = {Goodfellow, M. and Pirouz, T.},
  title = {NUMERICAL CLASSIFICATION OF SPOROACTINOMYCETES CONTAINING MESO-DIAMINOPIMELIC
	ACID IN THE CELL-WALL},
  journal = {Journal of General Microbiology},
  year = {1982},
  volume = {128},
  pages = {503-527},
  number = {MAR},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Ridell1983,
  author = {Ridell, M. and Goodfellow, M.},
  title = {NUMERICAL CLASSIFICATION OF MYCOBACTERIUM-FARCINOGENES, MYCOBACTERIUM-SENEGALENSE
	AND RELATED TAXA},
  journal = {Journal of General Microbiology},
  year = {1983},
  volume = {129},
  pages = {599-611},
  number = {MAR},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Goodfellow1987,
  author = {Goodfellow, M. and Harwood, C. R. and Nahaie, M. R.},
  title = {IMPACT OF PLASMIDS AND GENETIC CHANGE ON THE NUMERICAL CLASSIFICATION
	OF STAPHYLOCOCCI},
  journal = {Zentralblatt Fur Bakteriologie Mikrobiologie Und Hygiene Series a-Medical
	Microbiology Infectious Diseases Virology Parasitology},
  year = {1987},
  volume = {266},
  pages = {60-85},
  number = {1-2},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Elhag1988,
  author = {Elhag, K. M. and Senthilselvan, A.},
  title = {A SEROGROUPING SCHEME FOR THE STUDY OF THE EPIDEMIOLOGY OF BACTEROIDES-FRAGILIS},
  journal = {Journal of Medical Microbiology},
  year = {1988},
  volume = {27},
  pages = {199-205},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Mattioni2002,
  author = {Mattioni, C. and Casasoli, M. and Gonzalez, M. and Ipinza, R. and
	Villani, F.},
  title = {Comparison of ISSR and RAPD markers to characterize three Chilean
	Nothofagus species},
  journal = {Theoretical and Applied Genetics},
  year = {2002},
  volume = {104},
  pages = {1064-1070},
  number = {6-7},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Castro2002,
  author = {Castro, D. and Pujalte, M. J. and Lopez-Cortes, L. and Garay, E.
	and Borrego, J. J.},
  title = {Vibrios isolated from the cultured manila clam (Ruditapes philippinarum):
	numerical taxonomy and antibacterial activities},
  journal = {Journal of Applied Microbiology},
  year = {2002},
  volume = {93},
  pages = {438-447},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Hahm2003,
  author = {Hahm, B. K. and Maldonado, Y. and Schreiber, E. and Bhunia, A. K.
	and Nakatsu, C. H.},
  title = {Subtyping of foodborne and environmental isolates of Escherichia
	coli by multiplex-PCR, rep-PCR, PFGE, ribotyping and AFLP},
  journal = {Journal of Microbiological Methods},
  year = {2003},
  volume = {53},
  pages = {387-399},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Drossou2004,
  author = {Drossou, A. and Katsiotis, A. and Leggett, J. M. and Loukas, M. and
	Tsakas, S.},
  title = {Genome and species relationships in genus Avena based on RAPD and
	AFLP molecular markers},
  journal = {Theoretical and Applied Genetics},
  year = {2004},
  volume = {109},
  pages = {48-54},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Dimitriadou2004,
  author = {Dimitriadou, E. and Barth, M. and Windischberger, C. and Hornik,
	K. and Moser, E.},
  title = {A quantitative comparison of functional MRI cluster analysis},
  journal = {Artificial Intelligence in Medicine},
  year = {2004},
  volume = {31},
  pages = {57-71},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Hartwell2005,
  author = {Hartwell, S. I. and Claflin, L. W.},
  title = {Cluster analysis of contaminated sediment data: Nodal analysis},
  journal = {Environmental Toxicology and Chemistry},
  year = {2005},
  volume = {24},
  pages = {1816-1834},
  number = {7},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Deutsch2006,
  author = {Deutsch, R. and Cherner, M. and Grant, I.},
  title = {Significance testing of a cluster of multivariate binary variables:
	comparison of the tripartite T index to three common similarity measures},
  journal = {Statistical Methods in Medical Research},
  year = {2006},
  volume = {15},
  pages = {285-299},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@INPROCEEDINGS{Brickle2006,
  author = {Brickle, P. and MacKenzie, K.},
  title = {Parasites as biological tags for Eleginops maclovinus (Teleostei
	: Eleginopidae) around the Falkland Islands},
  booktitle = {11th International Congress of Parasitology (ICOPA XI)},
  year = {2006},
  pages = {147-153},
  address = {Glasgow, SCOTLAND},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Ercisli2008,
  author = {Ercisli, S. and Orhan, E. and Hizarci, Y. and Yildirim, N. and Agar,
	G.},
  title = {Genetic diversity in grapevine germplasm resources in the Coruh valley
	revealed by RAPD markers},
  journal = {Biochemical Genetics},
  year = {2008},
  volume = {46},
  pages = {590-597},
  number = {9-10},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Stenstrom1990,
  author = {Stenstrom, I. M. and Molin, G.},
  title = {CLASSIFICATION OF THE SPOILAGE FLORA OF FISH, WITH SPECIAL REFERENCE
	TO SHEWANELLA-PUTREFACIENS},
  journal = {Journal of Applied Bacteriology},
  year = {1990},
  volume = {68},
  pages = {601-618},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Molin1992,
  author = {Molin, G. and Johansson, M. L. and Stahl, M. and Ahrne, S. and Andersson,
	R. and Jeppsson, B. and Bengmark, S.},
  title = {SYSTEMATICS OF THE LACTOBACILLUS POPULATION ON RAT INTESTINAL-MUCOSA
	WITH SPECIAL REFERENCE TO LACTOBACILLUS-REUTERI},
  journal = {Antonie Van Leeuwenhoek International Journal of General and Molecular
	Microbiology},
  year = {1992},
  volume = {61},
  pages = {175-183},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Molin1993,
  author = {Molin, G. and Jeppsson, B. and Johansson, M. L. and Ahrne, S. and
	Nobaek, S. and Stahl, M. and Bengmark, S.},
  title = {Numerical taxonomy of lactobacillus spp associated with healthy and
	diseased mucosa of the human intestines},
  journal = {Journal of Applied Bacteriology},
  year = {1993},
  volume = {74},
  pages = {314-323},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Ternstrom1993,
  author = {Ternstrom, A. and Lindberg, A. M. and Molin, G.},
  title = {CLASSIFICATION OF THE SPOILAGE FLORA OF RAW AND PASTEURIZED BOVINE-MILK,
	WITH SPECIAL REFERENCE TO PSEUDOMONAS AND BACILLUS},
  journal = {Journal of Applied Bacteriology},
  year = {1993},
  volume = {75},
  pages = {25-34},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Odonnell1993,
  author = {Odonnell, A. G. and Falconer, C. and Goodfellow, M. and Ward, A.
	C. and Williams, E.},
  title = {BIOSYSTEMATICS AND DIVERSITY AMONGST NOVEL CARBOXYDOTROPHIC ACTINOMYCETES},
  journal = {Antonie Van Leeuwenhoek International Journal of General and Molecular
	Microbiology},
  year = {1993},
  volume = {64},
  pages = {325-340},
  number = {3-4},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Yohalem1994,
  author = {Yohalem, D. S. and Lorbeer, J. W.},
  title = {INTRASPECIFIC METABOLIC DIVERSITY AMONG STRAINS OF BURKHOLDERIA-CEPACIA
	ISOLATED FROM DECAYED ONIONS, SOILS, AND THE CLINICAL ENVIRONMENT},
  journal = {Antonie Van Leeuwenhoek International Journal of General and Molecular
	Microbiology},
  year = {1994},
  volume = {65},
  pages = {111-131},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Johansson1995,
  author = {Johansson, M. L. and Sanni, A. and Lonner, C. and Molin, G.},
  title = {PHENOTYPICALLY BASED TAXONOMY USING API 50CH OF LACTOBACILLI FROM
	NIGERIAN OGI, AND THE OCCURRENCE OF STARCH FERMENTING STRAINS},
  journal = {International Journal of Food Microbiology},
  year = {1995},
  volume = {25},
  pages = {159-168},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Ramser1996,
  author = {Ramser, J. and LopezPeralta, C. and Wetzel, R. and Weising, K. and
	Kahl, G.},
  title = {Genomic variation and relationships in aerial yam (Dioscorea bulbifera
	L) detected by random amplified polymorphic DNA},
  journal = {Genome},
  year = {1996},
  volume = {39},
  pages = {17-25},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.19}
}

@ARTICLE{Kim2006,
  author = {Kim, J. and Jacobs, D. R. and Luepker, R. V. and Shahar, E. and Margolis,
	K. L. and Becker, M. P.},
  title = {Prognostic value of a novel classification scheme for heart failure:
	The Minnesota Heart Failure Criteria},
  journal = {American Journal of Epidemiology},
  year = {2006},
  volume = {164},
  pages = {184-193},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.29}
}

@ARTICLE{King2007,
  author = {King, W. M. and Giess, S. A. and Lombardino, L. J.},
  title = {Subtyping of children with developmental dyslexia via bootstrap aggregated
	clustering and the gap statistic: comparison with the double-deficit
	hypothesis},
  journal = {International Journal of Language \& Communication Disorders},
  year = {2007},
  volume = {42},
  pages = {77-95},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.29}
}

@INPROCEEDINGS{Skeem2004,
  author = {Skeem, J. and Johansson, P. and Andershed, H. and Kerr, M. and Louden,
	J. E.},
  title = {Two subtypes of psychopathic violent offenders that parallel primary
	and secondary variants},
  booktitle = {4th Conference of the International-Association-of-Forensic-Mental-Health-Services},
  year = {2004},
  pages = {395-409},
  address = {Stockholm, SWEDEN},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.29}
}

@ARTICLE{Mun2008,
  author = {Mun, E. Y. and von Eye, A. and Bates, M. E. and Vaschillo, E. G.},
  title = {Finding groups using model-based cluster analysis: Heterogeneous
	emotional self-regulatory processes and heavy alcohol use risk},
  journal = {Developmental Psychology},
  year = {2008},
  volume = {44},
  pages = {481-495},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.29}
}

@ARTICLE{Munson2008,
  author = {Munson, J. and Dawson, G. and Sterling, L. and Beauchaine, T. and
	Zhou, A. and Koehler, E. and Lord, C. and Rogers, S. and Sigman,
	M. and Estes, A. and Abbott, R.},
  title = {Evidence for Latent Classes of IQ in Young Children With Autism Spectrum
	Disorder},
  journal = {American Journal on Mental Retardation},
  year = {2008},
  volume = {113},
  pages = {439-452},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.06.29}
}

@MISC{SASgeneral,
  author = {{SAS}},
  title = {Copyright, {SAS} {I}nstitute {I}nc. {SAS} and all other {SAS} {I}nstitute
	{I}nc. product or service names are registered trademarks or trademarks
	of {SAS} {I}nstitute {I}nc., {C}ary, {NC}, {USA}.},
  year = {2009},
  owner = {rebeccaspeckman},
  timestamp = {2009.07.04}
}

@BOOK{Maxcy1998,
  title = {Maxcy-Rosenau-Last public health \& preventive medicine},
  publisher = {Appleton \& Lange},
  year = {1998},
  author = {Maxcy, Kenneth Fuller and Rosenau, M. J. and Last, John M. and Wallace,
	Robert B.},
  address = {Stamford, Conn.},
  edition = {14th},
  note = {97053116 editor, Robert B. Wallace. Public health \& preventive medicine
	Public health and preventive medicine ill., maps ; 29 cm. Includes
	bibliographical references and index.},
  keywords = {Public health. Medicine, Preventive. Public Health Preventive Medicine},
  owner = {rebeccaspeckman},
  timestamp = {2009.07.04}
}

@BOOK{,
  owner = {rebeccaspeckman},
  timestamp = {2009.07.04}
}

@INCOLLECTION{Tyler1998,
  author = {Tyler, Carl W. and Last, John M.},
  title = {Epidemiology},
  booktitle = {Maxcy-Rosenau-Last public health \& preventive medicine},
  publisher = {Appleton \& Lange},
  year = {1998},
  editor = {Maxcy, Kenneth Fuller and Rosenau, M. J. and Last, John M. and Wallace,
	Robert B.},
  address = {Stamford, Conn.},
  edition = {14th},
  owner = {rebeccaspeckman},
  timestamp = {2009.07.04}
}

@INCOLLECTION{Milligan1996,
  author = {Milligan, Glenn W.},
  title = {Clustering validation: {R}esults and implications for applied analyses},
  booktitle = {Clustering and classification},
  publisher = {World Scientific},
  year = {1996},
  editor = {Arabie, Phipps and Hubert, Lawrence J. and Soete, Geert de},
  pages = {341--375},
  address = {Singapore; River Edge, NJ},
  keywords = {Cluster analysis. Discriminant analysis.},
  owner = {rebeccaspeckman},
  timestamp = {2009.07.04}
}

@ARTICLE{Nylund2007,
  author = {Nylund, K. L. and Asparoutiov, T. and Muthen, B. O.},
  title = {Deciding on the number of classes in latent class analysis and growth
	mixture modeling: A Monte Carlo simulation study},
  journal = {Structural Equation Modeling-a Multidisciplinary Journal},
  year = {2007},
  volume = {14},
  pages = {535-569},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.07.27}
}

@ARTICLE{Yang2007,
  author = {Yang, C. C. and Yang, C. C.},
  title = {Separating latent classes by information criteria},
  journal = {Journal of Classification},
  year = {2007},
  volume = {24},
  pages = {183-203},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.07.27}
}

@ARTICLE{Nylund2008,
  author = {Nylund, K. L. and Asparouhov, T. and Muthen, B. O.},
  title = {Deciding on the number of classes in latent class analysis and growth
	mixture modeling: A Monte Carlo simulation study (vol 14, pg 535,
	2007)},
  journal = {Structural Equation Modeling-a Multidisciplinary Journal},
  year = {2008},
  volume = {15},
  pages = {182-182},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.07.27}
}

@ARTICLE{Somers2002,
  author = {Somers, E. and Magder, L. S. and Petri, M.},
  title = {Antiphospholipid antibodies and incidence of venous thrombosis in
	a cohort of patients with systemic lupus erythematosus},
  journal = {Journal of Rheumatology},
  year = {2002},
  volume = {29},
  pages = {2531-2536},
  number = {12},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Smolen1998,
  author = {Smolen, J. S. and Steiner, G.},
  title = {Mixed connective tissue disease - To be or not to be?},
  journal = {Arthritis and Rheumatism},
  year = {1998},
  volume = {41},
  pages = {768-777},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Singh-Manoux2002,
  author = {Singh-Manoux, A. and Clarke, P. and Marmot, M.},
  title = {Multiple measures of socio-economic position and psychosocial health:
	proximal and distal measures},
  journal = {International Journal of Epidemiology},
  year = {2002},
  volume = {31},
  pages = {1192-1199},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Sharp1999,
  author = {Sharp, G. C. and Hoffman, R. W.},
  title = {Clinical, immunologic, and immunogenetic evidence that mixed connective
	tissue disease is a distinct entity: comment on the article by Smolen
	and Steiner},
  journal = {Arthritis and Rheumatism},
  year = {1999},
  volume = {42},
  pages = {190-191},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Schroeder2001,
  author = {Schroeder, J. C. and Weinberg, C. R.},
  title = {Use of missing-data methods to correct bias and improve precision
	in case-control studies in which cases are subtyped but subtype information
	is incomplete},
  journal = {American Journal of Epidemiology},
  year = {2001},
  volume = {154},
  pages = {954-962},
  number = {10},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Schoenfeld1983,
  author = {Schoenfeld, D. A.},
  title = {SAMPLE-SIZE FORMULA FOR THE PROPORTIONAL-HAZARDS REGRESSION-MODEL},
  journal = {Biometrics},
  year = {1983},
  volume = {39},
  pages = {499-503},
  number = {2},
  note = {ISI Document Delivery No.: RC899 Times Cited: 165 Cited Reference
	Count: 8},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Schafer2002,
  author = {Schafer, J. L. and Graham, J. W.},
  title = {Missing data: Our view of the state of the art},
  journal = {Psychological Methods},
  year = {2002},
  volume = {7},
  pages = {147-177},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Sarwal2003,
  author = {Sarwal, M. and Chua, M. S. and Kambham, N. and Hsieh, S. C. and Satterwhite,
	T. and Masek, M. and Salvatierra, O.},
  title = {Molecular heterogeneity in acute renal allograft rejection identified
	by DNA microarray profiling},
  journal = {New England Journal of Medicine},
  year = {2003},
  volume = {349},
  pages = {125-138},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Rothschild1983,
  author = {Rothschild, B. M. and Jones, J. V. and Chesney, C. and Pifer, D.
	D. and Thompson, L. D. and James, K. K. and Badger, H.},
  title = {RELATIONSHIP OF CLINICAL FINDINGS IN SYSTEMIC LUPUS-ERYTHEMATOSUS
	TO SERO-REACTIVITY},
  journal = {Arthritis and Rheumatism},
  year = {1983},
  volume = {26},
  pages = {45-51},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Ramos-Casals2007,
  author = {Ramos-Casals, M. and Brito-Zeron, P. and Font, J.},
  title = {The overlap of Sjogren's syndrome with other systemic autoimmune
	diseases},
  journal = {Seminars in Arthritis and Rheumatism},
  year = {2007},
  volume = {36},
  pages = {246-255},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Pulkstenis2004,
  author = {Pulkstenis, E. and Robinson, T. J.},
  title = {Goodness-of-fit tests for ordinal response regression models},
  journal = {Statistics in Medicine},
  year = {2004},
  volume = {23},
  pages = {999-1014},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Petri1991,
  author = {Petri, M. and Perezgutthann, S. and Longenecker, J. C. and Hochberg,
	M.},
  title = {MORBIDITY OF SYSTEMIC LUPUS-ERYTHEMATOSUS - ROLE OF RACE AND SOCIOECONOMIC-STATUS},
  journal = {American Journal of Medicine},
  year = {1991},
  volume = {91},
  pages = {345-353},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Petri1993,
  author = {Petri, M. and Caffentzis, E. and Conroy, M. and Goldman, D.},
  title = {Clinical presentation of systemic lupus erythematosus (SLE), 1960-92},
  journal = {Arthritis and Rheumatism},
  year = {1993},
  volume = {36},
  pages = {R22-R22},
  number = {5},
  note = {Suppl. S},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Petri2000,
  author = {Petri, M.},
  title = {Detection of coronary artery disease and the role of traditional
	risk factors in the Hopkins Lupus Cohort},
  journal = {Lupus},
  year = {2000},
  volume = {9},
  pages = {170-175},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Petri2000a,
  author = {Petri, M.},
  title = {Hopkins Lupus Cohort - 1999 update},
  journal = {Rheumatic Disease Clinics of North America},
  year = {2000},
  volume = {26},
  pages = {199-+},
  number = {2},
  abstract = {The Hopkins Lupus Cohort is a decade-long prospective study, now numbering
	800 patients with systemic lupus erythematosus. In this article,
	predictors of disease activity, disease damage (including accelerated
	atherosclerosis and antiphospholipid antibody syndrome) and health
	status are reviewed.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Petri1997,
  author = {Petri, M.},
  title = {Smoking is a risk factor for musculoskeletal. Pulmonary and cardiac
	disease in systemic lupus erythematosus},
  journal = {Arthritis and Rheumatism},
  year = {1997},
  volume = {40},
  pages = {527-527},
  number = {9},
  note = {Suppl. S},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Pepe2008,
  author = {Pepe, M. S. and Feng, Z. and Huang, Y. and Longton, G. and Prentice,
	R. and Thompson, I. M. and Zheng, Y.},
  title = {Integrating the predictiveness of a marker with its performance as
	a classifier},
  journal = {American Journal of Epidemiology},
  year = {2008},
  volume = {167},
  pages = {362-368},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Panush1993,
  author = {Panush, R. S. and Greer, J. M. and Morshedian, K. K.},
  title = {WHAT IS LUPUS - WHAT IS NOT LUPUS},
  journal = {Rheumatic Disease Clinics of North America},
  year = {1993},
  volume = {19},
  pages = {223-234},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Okamura1993,
  author = {Okamura, M. and Kanayama, Y. and Amastu, K. and Negoro, N. and Kohda,
	S. and Takeda, T. and Inoue, T.},
  title = {SIGNIFICANCE OF ENZYME-LINKED-IMMUNOSORBENT-ASSAY (ELISA) FOR ANTIBODIES
	TO DOUBLE-STRANDED AND SINGLE-STRANDED-DNA IN PATIENTS WITH LUPUS
	NEPHRITIS - CORRELATION WITH SEVERITY OF RENAL HISTOLOGY},
  journal = {Annals of the Rheumatic Diseases},
  year = {1993},
  volume = {52},
  pages = {14-20},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Obuchowski2004,
  author = {Obuchowski, N. A. and Lieber, M. L. and Wians, F. H.},
  title = {ROC curves in Clinical chemistry: Uses, misuses, and possible solutions},
  journal = {Clinical Chemistry},
  year = {2004},
  volume = {50},
  pages = {1118-1125},
  number = {7},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Muller1992,
  author = {Muller, H. G.},
  title = {GOODNESS-OF-FIT DIAGNOSTICS FOR REGRESSION-MODELS},
  journal = {Scandinavian Journal of Statistics},
  year = {1992},
  volume = {19},
  pages = {157-172},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Mukherjee2009,
  author = {Mukherjee, B. and Liu, I.},
  title = {A note on bias due to fitting prospective multivariate generalized
	linear models to categorical outcomes ignoring retrospective sampling
	schemes},
  journal = {Journal of Multivariate Analysis},
  year = {2009},
  volume = {100},
  pages = {459-472},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Mosca2004,
  author = {Mosca, M. and Baldini, C. and Bombardieri, S.},
  title = {Undifferentiated connective tissue diseases in 2004},
  journal = {Clinical and Experimental Rheumatology},
  year = {2004},
  volume = {22},
  pages = {S14-S18},
  number = {3},
  note = {Suppl. 33},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Miller1983,
  author = {Miller, M. H. and Urowitz, M. B. and Gladman, D. D.},
  title = {THE SIGNIFICANCE OF THROMBOCYTOPENIA IN SYSTEMIC LUPUS-ERYTHEMATOSUS},
  journal = {Arthritis and Rheumatism},
  year = {1983},
  volume = {26},
  pages = {1181-1186},
  number = {10},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Migliorini2005,
  author = {Migliorini, P. and Baldini, C. and Rocchi, V. and Bombardieri, S.},
  title = {Anti-Sm and anti-RNP antibodies},
  journal = {Autoimmunity},
  year = {2005},
  volume = {38},
  pages = {47-54},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{McCarthy1999,
  author = {McCarthy, N. and Giesecke, J.},
  title = {Case-case comparisons to study causation of common infectious diseases},
  journal = {International Journal of Epidemiology},
  year = {1999},
  volume = {28},
  pages = {764-768},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{McCarthy,
  author = {McCarthy},
  title = {The existence of MLE in binary logistic regression},
  journal = {Source unclear},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@BOOK{Maxcy1998a,
  title = {Maxcy-Rosenau-Last public health \& preventive medicine},
  publisher = {Appleton \& Lange},
  year = {1998},
  author = {Maxcy, Kenneth Fuller and Rosenau, M. J. and Last, John M. and Wallace,
	Robert B.},
  address = {Stamford, Conn.},
  edition = {14th},
  note = {97053116 editor, Robert B. Wallace. Public health \& preventive medicine
	Public health and preventive medicine ill., maps ; 29 cm. Includes
	bibliographical references and index.},
  keywords = {Public health. Medicine, Preventive. Public Health Preventive Medicine},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Gladman1997,
  author = {Gladman, D. D. and Urowitz, M. B. and Goldsmith, C. H. and Fortin,
	P. and Ginzler, E. and Gordon, C. and Hanly, J. G. and Isenberg,
	D. A. and Kalunian, K. and Nived, O. and Petri, M. and SanchezGuerrero,
	J. and Snaith, M. and Sturfelt, G.},
  title = {The reliability of the systemic Lupus International Collaborating
	Clinics American College of Rheumatology Damage Index in patients
	with Systemic Lupus Erythematosus},
  journal = {Arthritis and Rheumatism},
  year = {1997},
  volume = {40},
  pages = {809-813},
  number = {5},
  abstract = {Objective. To test the reliability of the Systemic Lupus International
	Collaborating Clinics/American College of Rheumatology (SLICC/ACR)
	Damage Indes and the Systemic Lupus Erythematosus Disease Activity
	Index (SLEDAI) in the assessment of patients,vith SLE. Methods. Ten
	patients with SLE, representing a spectrum of damage and activity,
	were included, Each patient was examined by 6 of 10 physicians from
	5 countries, representing 10 lupus clinics. The SLICC/ACR Damage
	Indes was used to assess accumulated damage, and the SLEDAI was used
	to assess disease activity, The order of the patients and physicians
	was randomized according to a Youden square design. Results. The
	SLICC/ACR Damage Index detected differences among patients (P <0.001).
	There was no detectable observer difference (P=0.933), and there
	was no order effect (P=0.261). Similar results were obtained with
	the SLEDAI. There was concordance in the SLICC/ACR Damage Index among
	observers, despite a wide spectrum of disease activity detected by
	the SLEDAI. Conclusion. Physicians from different centers are able
	to assess patients with SLE in a reproducible way, using the SLEDAI
	to assess disease activity and the SLICC/ACR Damage Index to assess
	accumulated damage.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Gladman2003,
  author = {Gladman, D. D. and Hirani, N. and Ibanez, D. and Urowitz, M. B.},
  title = {Clinically active serologically quiescent systemic lupus erythematosus},
  journal = {Journal of Rheumatology},
  year = {2003},
  volume = {30},
  pages = {1960-1962},
  number = {9},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Gladman1996,
  author = {Gladman, D. and Ginzler, E. and Goldsmith, C. and Fortin, P. and
	Liang, M. and Urowitz, M. and Bacon, P. and Bombardieri, S. and Hanly,
	J. and Hay, E. and Isenberg, D. and Jones, J. and Kalunian, K. and
	Maddison, P. and Nived, O. and Petri, M. and Richter, M. and SanchezGuerrero,
	J. and Snaith, M. and Sturfelt, G. and Symmons, D. and Zoma, A.},
  title = {The development and initial validation of the systemic lupus international
	collaborating clinics American College of Rheumatology Damage Index
	for Systemic Lupus Erythematosus},
  journal = {Arthritis and Rheumatism},
  year = {1996},
  volume = {39},
  pages = {363-369},
  number = {3},
  abstract = {Objective. To develop and perform an initial validation of a damage
	index for systemic lupus erythematosus (SLE). Methods. A list of
	items considered to reflect damage in SLE was generated through a
	nominal group process. A consensus as to which items to be included
	in an index was reached, together with rules for ascertainment. Each
	center submitted 2 assessments, 5 years apart, on 2 patients with
	active and 2 with inactive disease, of whom 1 had increased damage
	and the other had stable disease. Analysis of variance was used to
	test the factors physician, time, amount of damage, and activity
	status. Results. Nineteen physicians completed the damage index on
	42 case scenarios. The analysis revealed that the damage index could
	identify changes in damage seen in patients with both active and
	inactive disease. Patients who had active disease at both time points
	had a higher increase in damage. There was good agreement among the
	physicians on the assessment of damage in these patients. Conclusion.
	This damage index for SLE records damage occurring in patients with
	SLE regardless of its cause. The index was demonstrated to have content,
	face, criterion, and discriminant validity.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Garcia1996,
  author = {Garcia, C. O. and Molina, J. F. and GutierrezUrena, S. and Scopelitis,
	E. and Wilson, W. A. and Gharavi, A. E. and Espinoza, L. R.},
  title = {Autoantibody profile in African-American patients with lupus nephritis},
  journal = {Lupus},
  year = {1996},
  volume = {5},
  pages = {602-605},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Furnival1974,
  author = {Furnival, G. M. and Wilson, R. W.},
  title = {Regressions by leaps and bounds},
  journal = {Technometrics},
  year = {1974},
  volume = {16},
  pages = {499-511},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Font2004,
  author = {Font, J. and Cervera, R. and Ramos-Casals, M. and Garcia-Carrasco,
	M. and Sentis, J. and Herrero, C. and del Olmo, J. A. and Darnell,
	A. and Ingelmo, M.},
  title = {Clusters of clinical and immunologic features in systemic lupus erythematosus:
	Analysis of 600 patients from a single center},
  journal = {Seminars in Arthritis and Rheumatism},
  year = {2004},
  volume = {33},
  pages = {217-230},
  number = {4},
  abstract = {Objective: To analyze the prevalence and characteristics of the main
	clinical, hematologic, and immunologic manifestations of systemic
	lupus erythematosus (SLE) in a cohort of 600 consecutive patients
	from a single center, and to determine the specific characteristics
	of organ involvement in a homogeneous SLE population. Methods: Patients
	were consecutively seen in our department either as inpatients or
	outpatients between 1980 and 2001. All had documented medical histories
	and underwent a medical interview as well as a routine general physical
	examination. Clinical and serologic characteristics of all patients
	were consecutively collected in a protocol form. Results: The final
	cohort (survival cohort) consisted of 533 (89%) women and 67 (11%)
	men (female to male ratio, 8:1), with an average of 29 new patients
	per year. Mean age at onset of symptoms attributable to the disease
	was 31 years (range, 5 to 84 years) and mean age at the time of diagnosis
	of SLE was 33 years (range, 6 to 85 years). The most frequent SLE
	involvement was articular involvement, found in 498 patients (83%),
	followed by hematologic involvement in 451 patients (75%), specific
	SLE cutaneous involvement in 354 patients (59%), constitutional features
	in 252 patients (42%), and nephropathy in 203 patients (34%). Patients
	enrolled in the protocol before 1991 had a higher frequency of central
	nervous system (CNS) involvement (27% vs 10%, P < .001), thrombotic
	events (17% vs 9%, P = .003), and abnormal hematologic parameters
	(85% vs 66%, P < .01), but a lower frequency of articular involvement
	(79% vs 86%, P = .038) than those enrolled after 1991. The following
	were observed associations: specific SLE cutaneous involvement was
	associated with anti-Sm antibodies; renal involvement with hemolytic
	anemia and anti-double-sranded DNA antibodies; CNS involvement with
	thrombocytopenia and immunoglobulin G-anticardiolipin; thrombotic
	events with low total hemolytic complement, immunoglobulin G-anticardiolipin,
	and lupus anticoagulant; and myositis with anemia and anti-ribonucleoprotein
	antibodies. Conclusion: This large study, performed in a single center,
	has shown cluster associations between certain clinical, hematologic,
	and immunologic features of SLE, reflecting specific patterns of
	disease expression. The accurate evaluation of clinical features
	and laboratory markers at disease diagnosis and during evolution
	may improve the clinical treatment of SLE patients. (C) 2004 Elsevier
	Inc. All rights reserved.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Firth1993,
  author = {Firth, D.},
  title = {Bias reduction of maximum-likelihood estimates},
  journal = {Biometrika},
  year = {1993},
  volume = {80},
  pages = {27-38},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Fagerland2008,
  author = {Fagerland, M. W. and Hosmer, D. W. and Bofin, A. M.},
  title = {Multinomial goodness-of-fit tests for logistic regression models},
  journal = {Statistics in Medicine},
  year = {2008},
  volume = {27},
  pages = {4238-4253},
  number = {21},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Edworthy1988,
  author = {Edworthy, S. M. and Zatarain, E. and McShane, D. J. and Bloch, D.
	A.},
  title = {Analysis of the 1982 ARA lupus criteria data set by recursive partitioning
	methodology: new insights into the relative merit of individual criteria},
  journal = {Journal of Rheumatology},
  year = {1988},
  volume = {15},
  pages = {1493-1498},
  number = {10},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Doria2005,
  author = {Doria, A. and Mosca, M. and Gambari, P. F. and Bombardieri, S.},
  title = {Defining unclassifiable connective tissue diseases: Incomplete, undifferentiated,
	or both?},
  journal = {Journal of Rheumatology},
  year = {2005},
  volume = {32},
  pages = {213-215},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@BOOK{Mausner1985,
  title = {Epidemiology : an introductory text},
  publisher = {Saunders},
  year = {1985},
  author = {Mausner, Judith S. and Kramer, Shira and Bahn, Anita K.},
  address = {Philadelphia},
  edition = {2nd},
  note = {83020292 Judith S. Mausner, Shira Kramer ; with contributions by
	Peter Gann, G. Stephen Bowen ; with the collaboration of Richard
	Morton. ill. ; 23 cm. At head of title: Mausner \& Bahn. Includes
	bibliographies and index.},
  keywords = {Epidemiology.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Matheny2005,
  author = {Matheny, M. E. and Ohno-Machado, L. and Resnic, F. S.},
  title = {Discrimination and calibration of mortality risk prediction models
	in interventional cardiology},
  journal = {Journal of Biomedical Informatics},
  year = {2005},
  volume = {38},
  pages = {367-375},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Maddison2000,
  author = {Maddison, P. J.},
  title = {Mixed connective tissue disease: overlap syndromes},
  journal = {Best Practice \& Research in Clinical Rheumatology},
  year = {2000},
  volume = {14},
  pages = {111-124},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Lin2006,
  author = {Lin, H. Y. and Myers, L.},
  title = {Power and Type I error rates of goodness-of-fit statistics for binomial
	generalized estimating equations (GEE) models},
  journal = {Computational Statistics \& Data Analysis},
  year = {2006},
  volume = {50},
  pages = {3432-3448},
  number = {12},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Li2007,
  author = {Li, Q. Z. and Zhou, J. and Wandstrat, A. E. and Carr-Johnson, F.
	and Branch, V. and Karp, D. R. and Mohan, C. and Wakeland, E. K.
	and Olsen, N. J.},
  title = {Protein array autoantibody profiles for insights into systemic lupus
	erythematosus and incomplete lupus syndromes},
  journal = {Clinical and Experimental Immunology},
  year = {2007},
  volume = {147},
  pages = {60-70},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Lee2008,
  author = {Lee, S. S. and Singh, S. and Link, K. and Petri, M.},
  title = {High-sensitivity C-reactive protein as an associate of clinical subsets
	and organ damage in systemic lupus erythematosus},
  journal = {Seminars in Arthritis and Rheumatism},
  year = {2008},
  volume = {38},
  pages = {41-54},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Lasko2005,
  author = {Lasko, T. A. and Bhagwat, J. G. and Zou, K. H. and Ohno-Machado,
	L.},
  title = {The use of receiver operating characteristic curves in biomedical
	informatics},
  journal = {Journal of Biomedical Informatics},
  year = {2005},
  volume = {38},
  pages = {404-415},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Kokori2000,
  author = {Kokori, S. I. G. and Ioannidis, J. P. A. and Voulgarelis, M. and
	Tzioufas, A. G. and Moutsopoulos, H. M.},
  title = {Autoimmune hemolytic anemia in patients with systemic lupus erythematosus},
  journal = {American Journal of Medicine},
  year = {2000},
  volume = {108},
  pages = {198-204},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@BOOK{Kleinbaum2002,
  title = {Logistic regression : a self-learning text},
  publisher = {Springer},
  year = {2002},
  author = {Kleinbaum, David G. and Klein, Mitchel and Pryor, Erica Rihl},
  series = {Statistics for biology and health},
  address = {New York},
  edition = {2nd},
  note = {2002019728 David G. Kleinbaum, Mitchel Klein ; with contributions
	by Erica Rihl Pryor. ill. ; 25 cm. Includes bibliographical references
	(p. 503-505) and index.},
  keywords = {Medicine Research Statistical methods. Regression analysis. Logistic
	distribution.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Demidenko2007,
  author = {Demidenko, E.},
  title = {Sample size determination for logistic regression revisited},
  journal = {Statistics in Medicine},
  year = {2007},
  volume = {26},
  pages = {3385-3397},
  number = {18},
  abstract = {There is no consensus on the approach to compute the power and sample
	size with logistic regression. Some authors use the likelihood ratio
	test; some use the test on proportions; some suggest various approximations
	to handle the multivariate case. We advocate the use of the Wald
	test since the Z-score is routinely used for statistical significance
	testing of regression coefficients. The null-variance formula became
	popular from early studies, which contradicts modern software, which
	utilizes the method of maximum likelihood estimation (MLE), when
	the variance of the MLE is estimated at the MLE, not at the null.
	We derive general Wald-based power and sample size formulas for logistic
	regression and then apply them to binary exposure and confounder
	to obtain a closed-form expression. These formulas are applied to
	minimize the total sample size in a case-control study to achieve
	a given power by optimizing the ratio of controls to cases. Approximately,
	the optimal number of controls to cases is equal to the square root
	of the alternative odds ratio. Our sample size and power calculations
	can be carried out online at www.dartmouth.edu/ similar to eugened.
	Copyright (c) 2006 John Wiley & Sons, Ltd.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Costenbader2004,
  author = {Costenbader, K. H. and Kim, D. J. and Peerzada, J. and Lockman, S.
	and Nobles-Knight, D. and Petri, M. and Karlson, E. W.},
  title = {Cigarette smoking and the risk of systemic lupus erythematosus -
	A meta-analysis},
  journal = {Arthritis and Rheumatism},
  year = {2004},
  volume = {50},
  pages = {849-857},
  number = {3},
  abstract = {Objective. Existing studies present conflicting evidence for the role
	of cigarette smoking as a risk factor in the development of systemic
	lupus erythematosus (SLE). We performed an extensive search of the
	medical literature for all studies examining this relationship, and
	performed a meta-analysis to arrive at a more precise estimate of
	effect. Methods. We performed a computerized literature search for
	all studies (in all languages), using Medline and EMBASE (1966 to
	present) and the Cochrane Collaboration database, and completed hand
	searches of relevant bibliographies and abstracts of conference proceedings.
	Several investigators systematically extracted data from the relevant
	studies. Unpublished data were obtained from the author of one abstract.
	Studies were examined in aggregate for heterogeneity and publication
	bias. The relationships of current smoking and past smoking (prior
	to the onset of SLE) to development of SLE were analyzed separately.
	Results. Fifty-two studies were identified and chosen for detailed
	review. Of these, 9 (7 case-control and 2 cohort studies) were appropriate
	for inclusion in our meta-analyses. For current smokers compared
	with nonsmokers, the odds ratio (OR) for development of SLE was significantly
	elevated (OR 1.50, 95% confidence interval [95% CI] 1.09-2.08). Former
	smokers, compared with nonsmokers, did not demonstrate an increased
	risk of SLE (OR 0.98, 95% CI 0.75-1.27). Several subgroups were also
	analyzed. Conclusion. Our meta-analysis of the 7 existing case-control
	and 2 cohort studies revealed a small but statistically significant
	association between current smoking and development of SLE. However,
	no association between past smoking and development of SLE was observed.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Cook2008,
  author = {Cook, N. R.},
  title = {Statistical evaluation of prognostic versus diagnostic models: Beyond
	the ROC curve},
  journal = {Clinical Chemistry},
  year = {2008},
  volume = {54},
  pages = {17-23},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Cook2007,
  author = {Cook, N. R.},
  title = {Use and misuse of the receiver operating characteristic curve in
	risk prediction},
  journal = {Circulation},
  year = {2007},
  volume = {115},
  pages = {928-935},
  number = {7},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Chatterjee2004,
  author = {Chatterjee, N.},
  title = {A two-stage regression model for epidemiological studies with multivariate
	disease classification data},
  journal = {Journal of the American Statistical Association},
  year = {2004},
  volume = {99},
  pages = {127-138},
  number = {465},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Cervera1993,
  author = {Cervera, R. and Khamashta, M. A. and Font, J. and Sebastiani, G.
	D. and Gil, A. and Lavilla, P. and Domenech, I. and Aydintug, A.
	O. and Jedrykagoral, A. and Deramon, E. and Galeazzi, M. and Haga,
	H. J. and Mathieu, A. and Houssiau, F. and Ingelmo, M. and Hughes,
	G. R. V. and Cervera, R. and Sebastiani, G. D. and Font, J. and Khamashta,
	M. A. and Hughes, G. R. V. and Font, J. and Cervera, R. and Lopezsoto,
	A. and Vivancos, J. and Ingelmo, M. and Urbanomarquez, A. and Khamashta,
	M. A. and Vianna, J. and Hughes, G. R. V. and Gil, A. and Lavilla,
	P. and Pintado, V. and Lopezdupla, M. and Vazquez, J. J. and Sebastiani,
	G. D. and Deramon, E. and Camps, M. and Frutos, M. A. and Perello,
	I. and Santos, P. G. and Abarca, M. and Nebro, A. F. and Domenech,
	I. and Tokgoz, G. and Aydintug, A. O. and Jedrykagoral, A. and Maldykowa,
	H. and Chwalinskasadowska, H. and Galeazzi, M. and Haga, H. J. and
	Mathieu, A. and Houssiau, F.},
  title = {Systemic lupus erythematosus: clinical and immunological patterns
	of disease expression in a cohort of 1,000 patients},
  journal = {Medicine},
  year = {1993},
  volume = {72},
  pages = {113-124},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Calvo-Alen1995,
  author = {Calvo-Alen, J. and Bastian, H. M. and Straaton, K. V. and Burgard,
	S. L. and Mikhail, I. S. and Alarcon, G. S.},
  title = {Identification of patient subsets among those presumptively diagnosed
	with, referred, and/or followed up for systemic lupus erythematosus
	at a large teriary care center},
  journal = {Arthritis and Rheumatism},
  year = {1995},
  volume = {38},
  pages = {1475-1484},
  number = {10},
  abstract = {Objective, To identify different subsets of patients from a large
	tertiary care center who were presumptively referred for and/or diagnosed
	with systemic lupus erythematosus (SLE) (or followed up), Methods.
	All patients who were referred, followed up, and/or diagnosed with
	SLE at our center, who had disease duration of less than or equal
	to 5 years, and who resided in Alabama, were identified and their
	charts reviewed and abstracted, Results, Abstracted data were reviewed
	by 3 rheumatologists, and patients were assigned to 1 of 3 categories:
	1) SLE by the American College of Rheumatology (ACR; formerly, the
	American Rheumatism Association) criteria, 2) clinical SLE but not
	meeting 4 of the ACR criteria, or 3) fibromyalgia-like manifestations
	with antinuclear antibody (ANA) positivity, There were 90 patients
	in the first group (criteria), 22 in the second group (clinical),
	and 37 in the third group (fibromyalgia-like), Patients in all 3
	groups were predominantly women, Only 5% of the fibromyalgia-like
	group were African-American, compared with 55-65% for the other 2
	groups. Organ system involvement occurred with comparable frequency
	in the first 2 groups, but mucocutaneous and hematologic abnormalities
	were more frequent in the criteria group; in contrast, the patients
	with fibromyalgia-like symptoms primarily presented with arthralgias/
	myalgias, fatigue, depression, and sleep disturbances, as well as
	mucocutaneous manifestations, Conclusion, When the ACR criteria for
	SLE are used to determine eligibility for lupus studies, a group
	of patients with clinically unequivocal SLE are excluded, A group
	of patients with fibromyalgia-like manifestations, who test positive
	for ANA and differ clinically and sociodemographically from the patients
	in the other 2 groups, very likely do not belong within the spectrum
	of SLE.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Bull2002,
  author = {Bull, S. B. and Mak, C. and Greenwood, C. M. T.},
  title = {A modified score function estimator for multinomial logistic regression
	in small samples},
  journal = {Computational Statistics \& Data Analysis},
  year = {2002},
  volume = {39},
  pages = {57-74},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Bull2007,
  author = {Bull, S. B. and Lewinger, J. P. and Lee, S. S. F.},
  title = {Confidence intervals for multinomial logistic regression in sparse
	data},
  journal = {Statistics in Medicine},
  year = {2007},
  volume = {26},
  pages = {903-918},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Kim2004,
  author = {Kim, K. Y. and Kim, B. J. and Yi, G. S.},
  title = {Reuse of imputed data in microarray analysis increases imputation
	efficiency},
  journal = {Bmc Bioinformatics},
  year = {2004},
  volume = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Kim2007,
  author = {Kim, D. W. and Lee, K. Y. and Lee, K. H. and Lee, D.},
  title = {Towards clustering of incomplete microarray data without the use
	of imputation},
  journal = {Bioinformatics},
  year = {2007},
  volume = {23},
  pages = {107-113},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@INPROCEEDINGS{Kavanaugh1998,
  author = {Kavanaugh, A. and Tomar, R. and Reveille, J. and Solomon, D. H. and
	Homburger, H. A.},
  title = {Guidelines for clinical use of the antinuclear antibody test and
	tests for specific autoantibodies to nuclear antigens},
  booktitle = {32nd College-of-American-Pathologists Conference},
  year = {1998},
  pages = {71-81},
  address = {Chicago, Illinois},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Kamen2008,
  author = {Kamen, D. L. and Barron, M. and Parker, T. M. and Shaftman, S. R.
	and Bruner, G. R. and Aberle, T. and James, J. A. and Scofield, R.
	H. and Harley, J. B. and Gilkeson, G. S.},
  title = {Autoantibody prevalence and lupus characteristics in a unique African
	American population},
  journal = {Arthritis and Rheumatism},
  year = {2008},
  volume = {58},
  pages = {1237-1247},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Jury2001,
  author = {Jury, E. C. and D'Cruz, D. and Morrow, W. J. W.},
  title = {Autoantibodies and overlap syndromes in autoimmune rheumatic disease},
  journal = {Journal of Clinical Pathology},
  year = {2001},
  volume = {54},
  pages = {340-347},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Jurencak2009,
  author = {Jurencak, R. and Fritzler, M. and Tyrrell, P. and Hiraki, L. and
	Benseler, S. and Silverman, E.},
  title = {Autoantibodies in Pediatric Systemic Lupus Erythematosus: Ethnic
	Grouping, Cluster Analysis, and Clinical Correlations},
  journal = {Journal of Rheumatology},
  year = {2009},
  volume = {36},
  pages = {416-421},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Janssens2007,
  author = {Janssens, Acjw and Moonesinghe, R. and Yang, Q. and Steyerberg, E.
	W. and van Duijn, C. M. and Khoury, M. J.},
  title = {The impact of genotype frequencies on the clinical validity of genomic
	profiling for predicting common chronic diseases},
  journal = {Genetics in Medicine},
  year = {2007},
  volume = {9},
  pages = {528-535},
  number = {8},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Jakobsdottir2009,
  author = {Jakobsdottir, J. and Gorin, M. B. and Conley, Y. P. and Ferrell,
	R. E. and Weeks, D. E.},
  title = {Interpretation of Genetic Association Studies: Markers with Replicated
	Highly Significant Odds Ratios May Be Poor Classifiers},
  journal = {Plos Genetics},
  year = {2009},
  volume = {5},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Zweig1993,
  author = {Zweig, M. H. and Campbell, G.},
  title = {RECEIVER-OPERATING CHARACTERISTIC (ROC) PLOTS - A FUNDAMENTAL EVALUATION
	TOOL IN CLINICAL MEDICINE},
  journal = {Clinical Chemistry},
  year = {1993},
  volume = {39},
  pages = {561-577},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Isenberg1997,
  author = {Isenberg, D. A. and Garton, M. and Reichlin, M. W. and Reichlin,
	M.},
  title = {Long-term follow-up of autoantibody profiles in Black female lupus
	patients and clinical comparison with Caucasian and Asian patients},
  journal = {British Journal of Rheumatology},
  year = {1997},
  volume = {36},
  pages = {229-233},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Zonana-Nacach2000,
  author = {Zonana-Nacach, A. and Barr, S. G. and Magder, L. S. and Petri, M.},
  title = {Damage in systemic lupus erythematosus and its association with corticosteroids},
  journal = {Arthritis and Rheumatism},
  year = {2000},
  volume = {43},
  pages = {1801-1808},
  number = {8},
  abstract = {Objective. To evaluate the association between corticosteroid use
	and organ damage in patients with systemic lupus erythematosus (SLE).
	Method. The occurrence and date of organ damage, as measured by the
	Systemic Lupus International Collaborating Clinics/American College
	of Rheumatology Damage Index, were determined for 539 patients enrolled
	in the Hopkins Lupus Cohort Study. The risk of damage associated
	with the cumulative prednisone dose, high-dose prednisone (greater
	than or equal to 60 mg/day for greater than or equal to 2 months),
	and pulse methylprednisolone (1,000 mg intravenously for 1-3 days)
	was estimated using Cox proportional hazards regression analyses,
	controlling for age, race, and sex. Risk estimates for the cumulative
	prednisone dose were based on a reference dose of 36.5 gm (e.g.,
	10 mg of prednisone daily for 10 years [or equivalent]). Results.
	The cumulative prednisone dose was significantly associated with
	the development of osteoporotic fractures (relative risk [RR] 2.5,
	95% confidence interval [95%, CI] 1.7, 3.7), symptomatic coronary
	artery disease (RR 1.7, 95% CI 1.1, 2.5), and cataracts (RR 1.9,
	95% CI 1.4, 2.5). Each intravenous pulse was associated with a small
	increase in the risk of osteoporotic fractures (RR 1.3, 95% CI 1.0,
	1.8); however, this result failed to reach statistical significance
	(P = 0.07). Each 2-month exposure to high-dose prednisone was associated
	with a 1.2-fold increase in the risk of both avascular necrosis (95%
	CI 1,I, 1.4) and stroke (95% CI 1.0, 1.5). Conclusion. SLE patients
	receiving long-term prednisone therapy were at significant risk of
	morbidity due to permanent organ damage. Additional research is required
	to determine the relative contributions of SLE disease activity and
	corticosteroids to the pathogenesis of specific types of organ damage.
	Furthermore, new steroid-sparing therapies are needed in order to
	treat disease activity and minimize cumulative and high-dose prednisone
	exposure.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Isenberg1999,
  author = {Isenberg, D. and Black, C.},
  title = {Naming names! Comment on the article by Smolen and Steiner},
  journal = {Arthritis and Rheumatism},
  year = {1999},
  volume = {42},
  pages = {191-193},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Zabaleta-Lanz2006,
  author = {Zabaleta-Lanz, M. E. and Munoz, L. E. and Tapanes, F. J. and Vargas-Arenas,
	R. E. and Daboin, I. and Barrios, Y. and Pinto, J. A. and Bianco,
	N. E.},
  title = {Further description of early clinically silent lupus nephritis},
  journal = {Lupus},
  year = {2006},
  volume = {15},
  pages = {845-851},
  number = {12},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Werth2005,
  author = {Werth, V. P.},
  title = {Clinical manifestations of cutaneous lupus erythematosus},
  journal = {Autoimmunity Reviews},
  year = {2005},
  volume = {4},
  pages = {296-302},
  number = {5},
  abstract = {The skin findings seen in lupus erythematosus can present with either
	lupus-specific or lupus-nonspecific findings, with lupus-specific
	skin disease showing findings histopathologically distinct for cutaneous
	lupus erythematosus. Lupus-specific skin diseases include chronic
	cutaneous, subacute cutaneous, and acute cutaneous lupus erythematosus.
	The types of skin lesions in each group are clinically distinct and
	recognizing the specific subsets helps in prognosticating the likelihood
	of underlying systemic lupus. A number of medications are associated
	with cutaneous lupus, in particular with subacute cutaneous lupus
	erythematosus. Lupus nonspecific skin lesions are not histopathologically
	distinct for cutaneous lupus and/or may be seen as a feature of another
	disease process. Nonspecific disease-related skin lesions are frequently
	seen in patients with SLE, usually in the active phase of the disease.
	The current ACR classification criteria for SLE include four somewhat
	overlapping dermatologic criteria, butterfly rash, discoid lupus,
	photosensitivity, and oral ulcers and thus patients can be classified
	as having SLE with only skin manifestations. (c) 2005 Elsevier B.V.
	All rights reserved.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Bull1997,
  author = {Bull, S. B. and Greenwood, C. M. T. and Hauck, W. W.},
  title = {Jackknife bias reduction for polychotomous logistic regression},
  journal = {Statistics in Medicine},
  year = {1997},
  volume = {16},
  pages = {545-560},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Ward1990,
  author = {Ward, M. M. and Studenski, S.},
  title = {CLINICAL MANIFESTATIONS OF SYSTEMIC LUPUS-ERYTHEMATOSUS - IDENTIFICATION
	OF RACIAL AND SOCIOECONOMIC INFLUENCES},
  journal = {Archives of Internal Medicine},
  year = {1990},
  volume = {150},
  pages = {849-853},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@BOOK{Brownson2006,
  title = {Applied epidemiology : theory to practice},
  publisher = {Oxford University Press},
  year = {2006},
  author = {Brownson, Ross C. and Petitti, Diana B.},
  address = {Oxford ; New York},
  edition = {2nd},
  note = {2005053914 edited by Ross C. Brownson, Diana B. Petitti. ill. ; 25
	cm. Includes bibliographical references and index. Epidemiology:
	a foundation of public health -- Public health surveillance -- Outbreak
	and cluster investigations -- Systematic reviews in public health
	-- Epidemiology and risk assessment -- Epidemiologic issues in community
	intervention -- Outcomes research -- Measuring the quality of health
	care -- Epidemiology and health policy -- Epidemiology and the law
	-- Communicating epidemiologic information.},
  keywords = {Epidemiology Research Methodology. Epidemiology. Epidemiologic Methods.
	Health Policy. Quality of Health Care.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Ward1995,
  author = {Ward, M. M. and Pyun, E. and Studenski, S.},
  title = {LONG-TERM SURVIVAL IN SYSTEMIC LUPUS-ERYTHEMATOSUS - PATIENT CHARACTERISTICS
	ASSOCIATED WITH POORER OUTCOMES},
  journal = {Arthritis and Rheumatism},
  year = {1995},
  volume = {38},
  pages = {274-283},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@BOOK{Brownson1998,
  title = {Applied epidemiology : theory to practice},
  publisher = {Oxford University Press},
  year = {1998},
  author = {Brownson, Ross C. and Petitti, Diana B.},
  address = {New York},
  note = {97016894 edited by Ross C. Brownson, Diana B. Petitti. ill. ; 24
	cm. Includes bibliographical references and index.},
  keywords = {Epidemiology Research Methodology. Epidemiology. Epidemiologic Methods.
	Quality of Health Care. Health Policy.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Vila2004,
  author = {Vila, L. M. and Alarcon, G. S. and McGwin, G. and Friedman, A. W.
	and Baethge, B. A. and Bastian, H. M. and Fessler, B. J. and Reveille,
	J. D.},
  title = {Early clinical manifestations, disease activity and damage of systemic
	lupus erythematosus among two distinct US Hispanic subpopulations},
  journal = {Rheumatology},
  year = {2004},
  volume = {43},
  pages = {358-363},
  number = {3},
  abstract = {Objectives. To compare the baseline clinical manifestations, immunological
	features, disease activity and damage accrual in systemic lupus erythematosus
	(SLE) patients from two US Hispanic subgroups. Methods. A total of
	105 Hispanic SLE patients from Texas (a population of Mexican or
	Central American ancestry) and 81 from the island of Puerto Rico
	(all Puerto Ricans) participating in a longitudinal study of outcome
	were examined. The socio-economic/demographic, clinical and immunological
	variables were obtained at the time of enrolment (T-0). Disease activity
	was determined with the Systemic Lupus Activity Measure (SLAM), and
	disease damage with the Systemic Lupus International Collaborating
	Clinics (SLICC) Damage Index (SDI). Disease activity was also determined
	at the time of diagnosis (T-D). Results. At T-0 Hispanics from Texas
	were younger than those from Puerto Rico (33.1 +/- 12.0 vs 37.5 +/-
	11.6 yr, P = 0.0125). Both groups were similar with regard to gender
	distribution (92.4 vs 95.1% females) and disease duration (1.4 +/-
	1.4 vs 1.7 +/- 1.3 yr). Hispanics from Texas were more likely to
	have serositis (60.0 vs 8.6%, P < 0.0001), renal involvement (41.0
	vs 13.6%, P < 0.0001), psychosis (5.7 vs 0.0%, P = 0.0365) and thrombocytopenia
	(21.0 vs 3.7%, P = 0.0006). On the other hand, Hispanics from Puerto
	Rico were more likely to have photosensitivity (81.5 vs 41.0%, P
	< 0.0001), malar rash (65.4 vs 45.7%, P = 0.0074) and discoid rash
	(13.6 vs 2.9%, P = 0.0060). At baseline, the presence of anti-dsDNA
	antibodies was higher in Hispanics from Texas (69.5% vs 46.9%, P
	= 0.0018) while anti-Ro antibodies were more frequent in Hispanics
	from Puerto Rico (24.7 vs 11.4%, P = 0.0175). Mean SLAM scores at
	T-D (12.9 +/- 6.4 vs 9.1 +/- 4.6, P < 0.0001) and T-0 (10.9 +/- 6.3
	vs 6.6 +/- 3.8, P < 0.0001) were significantly higher in Hispanics
	from Texas. Similarly, mean SDI scores at T-0 were higher in Hispanics
	from Texas (0.67 +/- 1.08 vs 0.26 +/- 0.54, P = 0.0026). By stepwise
	Poisson regression, SDI scores were associated with older age, disease
	activity and ethnicity (Hispanics from Texas). Conclusions. Early
	in SLE, marked differences are observed between Hispanics from Texas
	and Puerto Rico. Higher disease activity, more major organ involvement,
	higher frequency of anti-dsDNA antibodies and more damage accrual
	occur in Hispanic lupus patients from Texas than in those from Puerto
	Rico.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Bombardier1992,
  author = {Bombardier, C. and Gladman, D. D. and Urowitz, M. B. and Caron, D.
	and Chang, C. H.},
  title = {DERIVATION OF THE SLEDAI - A DISEASE-ACTIVITY INDEX FOR LUPUS PATIENTS},
  journal = {Arthritis and Rheumatism},
  year = {1992},
  volume = {35},
  pages = {630-640},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Biesheuvel2008,
  author = {Biesheuvel, C. J. and Vergouwe, Y. and Steyerberg, E. W. and Grobbee,
	D. E. and Moons, K. G. M.},
  title = {Polytomous logistic regression analysis could be applied more often
	in diagnostic research},
  journal = {Journal of Clinical Epidemiology},
  year = {2008},
  volume = {61},
  pages = {125-134},
  number = {2},
  abstract = {Objective: Physicians commonly consider the presence of all differential
	diagnoses simultaneously. Polytomous logistic regression modeling
	allows for simultaneous estimation of the probability of multiple
	diagnoses. We discuss and (empirically) illustrate the value of this
	method for diagnostic research. Study Design and Setting: We used
	data from a study on the diagnosis of residual retroperitoneal mass
	histology in patients presenting with nonseminomatous testicular
	germ cell tumor. The differential diagnoses include benign tissue,
	mature teratoma, and viable cancer. Probabilities of each diagnosis
	were estimated with a polytomous logistic regression model and compared
	with the probabilities estimated from two consecutive dichotomous
	logistic regression models. Results: We provide interpretations of
	the odds ratios derived from the polytomous regression model and
	present a simple score chart to facilitate calculation of predicted
	probabilities from the polytomous model. For both modeling methods,
	we show the calibration plots and receiver operating characteristics
	curve (ROC) areas comparing each diagnostic outcome category with
	the other two. The ROC areas for benign tissue, mature teratoma,
	and viable cancer were similar for both modeling methods, 0.83 (95%
	confidence interval [CI] = 0.80-0.85) vs. 0.83 (95% CI = 0.80-0.85),
	0.78 (95% CI = 0.75-0.81) vs. 0.78 (95% CI = 0.75-0.81), and 0.66
	(95% CI = 0.61-0.71) vs. 0.64 (95% CI = 0.59-0.69), for polytomous
	and dichotomous regression models, respectively. Conclusion: Polytomous
	logistic regression is a useful technique to simultaneously model
	predicted probabilities of multiple diagnostic outcome categories.
	The performance of a polytomous prediction model can be assessed
	similarly to a dichotomous logistic regression model, and predictions
	by a polytomous model can be made with a user-friendly method. Because
	the simultaneous consideration of the presence of multiple (differential)
	conditions serves clinical practice better than consideration of
	the presence of only one target condition, polytomous logistic regression
	could be applied more often in diagnostic research. (C) 2008 Elsevier
	Inc. All rights reserved.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Begg1994,
  author = {Begg, C. B. and Zhang, Z. F.},
  title = {STATISTICAL-ANALYSIS OF MOLECULAR EPIDEMIOLOGY STUDIES EMPLOYING
	CASE-SERIES},
  journal = {Cancer Epidemiology Biomarkers \& Prevention},
  year = {1994},
  volume = {3},
  pages = {173-175},
  number = {2},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Aringer2005,
  author = {Aringer, M. and Steiner, G. and Smolen, J. S.},
  title = {Does mixed connective tissue disease exist? Yes},
  journal = {Rheumatic Disease Clinics of North America},
  year = {2005},
  volume = {31},
  pages = {411-+},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Aringer2007,
  author = {Aringer, M. and Smolen, J. S.},
  title = {Mixed connective tissue disease: what is behind the curtain?},
  journal = {Best Practice \& Research in Clinical Rheumatology},
  year = {2007},
  volume = {21},
  pages = {1037-1049},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Arbuckle2003,
  author = {Arbuckle, M. R. and McClain, M. T. and Rubertone, M. V. and Scofield,
	R. H. and Dennis, G. J. and James, J. A. and Harley, J. B.},
  title = {Development of autoantibodies before the clinical onset of systemic
	lupus erythematosus},
  journal = {New England Journal of Medicine},
  year = {2003},
  volume = {349},
  pages = {1526-1533},
  number = {16},
  abstract = {BACKGROUND: Although much is known about the natural history of systemic
	lupus erythematosus (SLE), the development of SLE autoantibodies
	before the diagnosis of the disease has not been extensively explored.
	We investigated the onset and progression of autoantibody development
	before the clinical diagnosis. METHODS: The Department of Defense
	Serum Repository contains approximately 30 million specimens prospectively
	collected from more than 5 million U.S. Armed Forces personnel. We
	evaluated serum samples obtained from 130 persons before they received
	a diagnosis of SLE, along with samples from matched controls. RESULTS:
	In 115 of the 130 patients with SLE (88 percent), at least one SLE
	autoantibody tested was present before the diagnosis (up to 9.4 years
	earlier; mean, 3.3 years). Antinuclear antibodies were present in
	78 percent (at a dilution of 1:120 or more), anti-double-stranded
	DNA antibodies in 55 percent, anti-Ro antibodies in 47 percent, anti-La
	antibodies in 34 percent, anti-Sm antibodies in 32 percent, anti-nuclear
	ribonucleoprotein antibodies in 26 percent, and antiphospholipid
	antibodies in 18 percent. Antinuclear, antiphospholipid antibodies,
	anti-Ro, and anti-La antibodies were present earlier than anti-Sm
	and anti-nuclear ribonucleoprotein antibodies (a mean of 3.4 years
	before the diagnosis vs. 1.2 years, P=0.005). Anti-double-stranded
	DNA antibodies, with a mean onset 2.2 years before the diagnosis,
	were found later than antinuclear antibodies (P=0.06) and earlier
	than anti-nuclear ribonucleoprotein antibodies (P=0.005). For many
	patients, the earliest available serum sample was positive; therefore,
	these measures of the average time from the first positive antibody
	test to the diagnosis are underestimates of the time from the development
	of antibodies to the diagnosis. Of the 130 initial matched controls,
	3.8 percent were positive for one or more autoantibodies. CONCLUSIONS:
	Autoantibodies are typically present many years before the diagnosis
	of SLE. Furthermore, the appearance of autoantibodies in patients
	with SLE tends to follow a predictable course, with a progressive
	accumulation of specific autoantibodies before the onset of SLE,
	while patients are still asymptomatic.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Arbuckle2003a,
  author = {Arbuckle, M. R. and James, J. A. and Dennis, G. J. and Rubertone,
	M. V. and McClain, M. T. and Kim, X. R. and Harley, J. B.},
  title = {Rapid clinical progression to diagnosis among African-American men
	with systemic lupus erythematosus},
  journal = {Lupus},
  year = {2003},
  volume = {12},
  pages = {99-106},
  number = {2},
  abstract = {The initial clinical course of systemic lupus erythematosus (SLE)
	is variable, ranging from relatively minor manifestations progressing
	over years to rapid onset of fulminate disease. We sought to identify
	factors associated with the rapid manifestation of SLE. Chart review
	of military medical records was used to identify 130 patients who
	met the American College of Rheumatology classification criteria
	for SLE. Demographics, clinical criteria date of occurrence, and
	the date of SLE classification (at least four clinical criteria)
	met were documented. Prospectively stored serum samples prior to
	the diagnosis were evaluated for SLE autoantibodies. Median time
	from the first recorded criteria to diagnosis was significantly shorter
	in African-American (AA) males compared with AA females and European
	American (EA) females and males combined. AA males were more likely
	to have nephritis as their first clinical symptom. Also, less time
	transpired between the first clinical criterion and SLE diagnosis
	in AA males with nephritis than in other groups presenting with nephritis.
	Even when cases presenting with nephritis were excluded, a diagnosis
	of SLE was made more rapidly in AA males. African-American men progress
	from initial clinical manifestations to SLE diagnosis more rapidly
	than other ethnic or gender groups.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Hoffman2004,
  author = {Hoffman, I. E. A. and Peene, I. and Meheus, L. and Huizinga, T. W.
	J. and Cebecauer, L. and Isenberg, D. and De Bosschere, K. and Hulstaert,
	F. and Veys, E. M. and De Keyser, F.},
  title = {Specific antinuclear antibodies are associated with clinical features
	in systemic lupus erythematosus},
  journal = {Annals of the Rheumatic Diseases},
  year = {2004},
  volume = {63},
  pages = {1155-1158},
  number = {9},
  abstract = {Objectives: To study associations between antinuclear antibodies (ANA)
	and signs/symptoms in patients with systemic lupus erythematosus
	(SLE). Methods: A consecutive cohort of 289 patients with SLE was
	included; 235 fulfilled ACR criteria for SLE and were further analysed.
	ANA profiles were determined by line immunoassay and by indirect
	immunofluorescence on Crithidia luciliae. An extensive list of signs/symptoms
	was evaluated. Results: Five clusters of antibodies were defined
	by cluster analysis: 1 - antibodies to SmB, SmD, RNP-A, RNP-C, and
	RNP-70k; 2 - antibodies to Ro52, Ro60, and SSB; 3, 4, and 5 - antibodies
	to ribosomal P, histones and dsDNA, respectively. Significant associations
	( pless than or equal to 0.01) were found between anti-RNP-70k, anti-RNP-A,
	anti-RNP-C and Raynaud's phenomenon, between anti-RNP-A, anti-RNP-70k
	and leucopenia, and between anti-RNP-A, anti-RNP-C and a lower prevalence
	of urine cellular casts. Anti-SSA, anti-SSB were associated with
	xerostomia, and anti-SSB with pericarditis. Antibodies to ribosomal
	P were associated with haemolytic anaemia, leucopenia, and alopecia.
	Patients with anti-dsDNA antibodies had a higher risk for cellular
	casts and a lower risk for photosensitivity. Antihistone antibodies
	were associated with arthritis. Conclusions: In a large and consecutive
	cohort of patients with SLE, clusters of antibodies were identified.
	Previously reported associations of antibodies with symptoms were
	confirmed and new associations found.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Hochberg1985,
  author = {Hochberg, M. C. and Boyd, R. E. and Ahearn, J. M. and Arnett, F.
	C. and Bias, W. B. and Provost, T. T. and Stevens, M. B.},
  title = {SYSTEMIC LUPUS-ERYTHEMATOSUS - A REVIEW OF CLINICO-LABORATORY FEATURES
	AND IMMUNOGENETIC MARKERS IN 150 PATIENTS WITH EMPHASIS ON DEMOGRAPHIC
	SUBSETS},
  journal = {Medicine},
  year = {1985},
  volume = {64},
  pages = {285-295},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Hochberg1997,
  author = {Hochberg, M. C.},
  title = {Updating the American College of Rheumatology revised criteria for
	the classification of systemic lupus erythematosus},
  journal = {Arthritis and Rheumatism},
  year = {1997},
  volume = {40},
  pages = {1725-1725},
  number = {9},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Hermann2005,
  author = {Hermann, C. and Westergaard, T. and Pedersen, B. V. and Wohlfahrt,
	J. and Host, A. and Melbye, M.},
  title = {A comparison of risk factors for wheeze and recurrent cough in preschool
	children},
  journal = {American Journal of Epidemiology},
  year = {2005},
  volume = {162},
  pages = {345-350},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Venables2006,
  author = {Venables, P. J. W.},
  title = {Mixed connective tissue disease},
  journal = {Lupus},
  year = {2006},
  volume = {15},
  pages = {132-137},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@INPROCEEDINGS{Heinze2005,
  author = {Heinze, G.},
  title = {A comparative investigation of methods for logistic regression with
	separated or nearly separated data},
  booktitle = {26th Annual Meeting of the International-Society-for-Clinical-Biostatistics
	(ISCB26)},
  year = {2005},
  pages = {4216-4226},
  address = {Szeged, HUNGARY},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@INCOLLECTION{Tyler1998a,
  author = {Tyler and Last, John M.},
  title = {Epidemiology},
  booktitle = {Maxcy-Rosenau-Last public health \& preventive medicine},
  publisher = {Appleton \& Lange},
  year = {1998},
  editor = {Maxcy, Kenneth Fuller and Rosenau, M. J. and Last, John M. and Wallace,
	Robert B.},
  pages = {xxviii, 1291 p.},
  address = {Stamford, Conn.},
  edition = {14th},
  note = {97053116 editor, Robert B. Wallace. Public health \& preventive medicine
	Public health and preventive medicine ill., maps ; 29 cm. Includes
	bibliographical references and index.},
  keywords = {Public health. Medicine, Preventive. Public Health Preventive Medicine},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Hardle1993,
  author = {Hardle, W. and Mammen, E.},
  title = {COMPARING NONPARAMETRIC VERSUS PARAMETRIC REGRESSION FITS},
  journal = {Annals of Statistics},
  year = {1993},
  volume = {21},
  pages = {1926-1947},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@INBOOK{Torrence1997,
  chapter = {Epidemiologic concepts of disease},
  title = {Understanding epidemiology},
  publisher = {Mosby},
  year = {1997},
  author = {Torrence, Mary E.},
  series = {Mosby's biomedical science series},
  address = {St. Louis},
  note = {96039467 Mary E. Torrence. ill., maps ; 24 cm. Includes bibliographical
	references and index.},
  keywords = {Epidemiology. Epidemiology.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Greer1989,
  author = {Greer, J. M. and Panush, R. S.},
  title = {INCOMPLETE LUPUS-ERYTHEMATOSUS},
  journal = {Archives of Internal Medicine},
  year = {1989},
  volume = {149},
  pages = {2473-2476},
  number = {11},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{To2005,
  author = {To, C. H. and Petri, M.},
  title = {Is antibody clustering predictive of clinical subsets and damage
	in systemic lupus erythematosus?},
  journal = {Arthritis and Rheumatism},
  year = {2005},
  volume = {52},
  pages = {4003-4010},
  number = {12},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Greenland1999,
  author = {Greenland, S.},
  title = {A unified approach to the analysis of case-distribution (case-only)
	studies},
  journal = {Statistics in Medicine},
  year = {1999},
  volume = {18},
  pages = {1-15},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Tikly1996,
  author = {Tikly, M. and Burgin, S. and Mohanlal, P. and Bellingan, A. and George,
	J.},
  title = {Autoantibodies in black South Africans with systemic lupus erythematosus:
	Spectrum and clinical associations},
  journal = {Clinical Rheumatology},
  year = {1996},
  volume = {15},
  pages = {261-265},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Goeman2006,
  author = {Goeman, J. J. and le Cessie, S.},
  title = {A goodness-of-fit test for multinomial logistic regression},
  journal = {Biometrics},
  year = {2006},
  volume = {62},
  pages = {980-985},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Terborg1990,
  author = {Terborg, E. J. and Horst, G. and Hummel, E. J. and Limburg, P. C.
	and Kallenberg, C. G. M.},
  title = {MEASUREMENT OF INCREASES IN ANTI-DOUBLE-STRANDED DNA ANTIBODY-LEVELS
	AS A PREDICTOR OF DISEASE EXACERBATION IN SYSTEMIC LUPUS-ERYTHEMATOSUS
	- A LONG-TERM, PROSPECTIVE-STUDY},
  journal = {Arthritis and Rheumatism},
  year = {1990},
  volume = {33},
  pages = {634-643},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Terborg1990a,
  author = {Terborg, E. J. and Groen, H. and Horst, G. and Limburg, P. C. and
	Wouda, A. A. and Kallenberg, C. G. M.},
  title = {CLINICAL ASSOCIATIONS OF ANTIRIBONUCLEOPROTEIN ANTIBODIES IN PATIENTS
	WITH SYSTEMIC LUPUS-ERYTHEMATOSUS},
  journal = {Seminars in Arthritis and Rheumatism},
  year = {1990},
  volume = {20},
  pages = {164-173},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Ananth1997,
  author = {Ananth, C. V. and Kleinbaum, D. G.},
  title = {Regression models for ordinal responses: A review of methods and
	applications},
  journal = {International Journal of Epidemiology},
  year = {1997},
  volume = {26},
  pages = {1323-1333},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Tapanes2000,
  author = {Tapanes, F. J. and Vasquez, M. and Ramirez, R. and Matheus, C. and
	Rodriguez, M. A. and Bianco, N.},
  title = {Cluster analysis of antinuclear autoantibodies in the prognosis of
	SLE nephropathy: are anti-extractable nuclear antibodies protective?},
  journal = {Lupus},
  year = {2000},
  volume = {9},
  pages = {437-444},
  number = {6},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Allison2003,
  author = {Allison, P. D.},
  title = {Missing data techniques for structural equation modeling},
  journal = {Journal of Abnormal Psychology},
  year = {2003},
  volume = {112},
  pages = {545-557},
  number = {4},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Swanton2005,
  author = {Swanton, J. and Isenberg, D.},
  title = {Mixed connective tissue disease: Still crazy after all these years},
  journal = {Rheumatic Disease Clinics of North America},
  year = {2005},
  volume = {31},
  pages = {421-+},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Alarcon2004,
  author = {Alarcon, G. S. and McGwin, G. and Roseman, J. M. and Uribe, A. and
	Fessler, B. J. and Bastian, H. M. and Friedman, A. W. and Baethge,
	B. and Vila, L. M. and Reveille, J. D.},
  title = {Systemic lupus erythematosus in three ethnic groups. XIX. Natural
	history of the accrual of the American College of Rheumatology criteria
	prior to the occurrence of Criteria diagnosis},
  journal = {Arthritis \& Rheumatism-Arthritis Care \& Research},
  year = {2004},
  volume = {51},
  pages = {609-615},
  number = {4},
  abstract = {Objective. To determine how the American College of Rheumatology (ACR)
	criteria for the classification of systemic lupus erythematosus (SLE)
	accrue in a multiethnic cohort of SLE patients. Methods. SLE patients
	enrolled in a longitudinal study of outcome were analyzed (LUMINA;
	Lupus in Minorities: Nature versus nurture) for the manner in which
	ACR criteria manifestations occurred prior to the accrual of 4 of
	them. Time at which a criterion was said to be present was determined
	by review of all previously available medical records and interview.
	Univariable and multivariable Cox proportional hazard models were
	examined for the association with time to accrual of 4 ACR criteria;
	results were reported as hazard ratios. Results. There were 103 Texas
	Hispanic (of Mexican or Central America ancestry) patients, 55 Puerto
	Rico Hispanics, 176 African Americans, and 137 Caucasians. The mean
	+/- SD and median (range) time to accrual of 4 ACR criteria were
	29.4 +/- 52.0 months and 9.1 (0 - 328.7) months; time was shortest
	for the Texas Hispanics (18.4 +/- 42.8 and 5.0 [0 - 248] months)
	and longest for the Caucasians (39.9 +/- 59.3 months and 17.7 [0
	- 324.6] months). Arthritis was the most frequent first criterion
	(34.5%); it was followed by photosensitivity (18.8%). When 2 criteria
	occurred from the outset, the most frequent combination was arthritis
	and antinuclear antibody positivity followed by malar rash and photosensitivity.
	A Cox-regression multivariable model identified Hispanic ethnicity
	(from Texas) and HLA-DRB1*0301 as predictors of short time to criteria
	accrual, whereas older age and married/living together were associated
	with long time to criteria accrual. Conclusion. Significant variability
	in the evolution of ACR criteria manifestations does occur. Texas
	Hispanics are more likely to have a rapid evolution of criteria manifestations,
	but several years may elapse before ACR criteria are accrued.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Staniswalis1991,
  author = {Staniswalis, J. G. and Severini, T. A.},
  title = {DIAGNOSTICS FOR ASSESSING REGRESSION-MODELS},
  journal = {Journal of the American Statistical Association},
  year = {1991},
  volume = {86},
  pages = {684-692},
  number = {415},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Alarcon1999,
  author = {Alarcon, G. S. and Friedman, A. W. and Straaton, K. V. and Moulds,
	J. M. and Lisse, J. and Bastian, H. M. and McGwin, G. and Bartolucci,
	A. A. and Roseman, J. M. and Reveille, J. D.},
  title = {Systemic lupus erythematosus in three ethnic groups: III A comparison
	of characteristics early in the natural history of the LUMINA cohort},
  journal = {Lupus},
  year = {1999},
  volume = {8},
  pages = {197-209},
  number = {3},
  abstract = {Aim: To determine and contrast the socioeconomic-demographic and clinical
	features of patients with recent onset (less than or equal to 5 y)
	systemic lupus erythematosus (SLE) from three ethnic groups, Hispanic,
	African-American and Caucasian (H, AA, C). Subjects and methods:
	SLE cases (American College of Rheumatology criteria) (incident (n
	= 56), prevalent (n = 173)), were enrolled in a longitudinal study
	at The University of Alabama at Birmingham, The University of Texas-Houston
	Health Science Center and The University of Texas Medical Branch
	at Galveston. Socioeconomic-demographic, clinical, immunological,
	behavioral and psychological data were obtained using validated instruments
	and standard laboratory techniques, and compared. Results: 70 H,
	88 AA and 71 C SLE patients constitute this cohort. H and AA patients
	were younger and of lower sacioeconomic-demographic status. They
	also had evidence of more frequent organ system involvement (renal,
	cardiovascular), more auto-antibodies, more active disease (after
	adjusting for discrepant socioeconomic-demographic features), lower
	levels of social support and more abnormal illness-related behaviors
	(more in H than in AA). H also were more likely to have an abrupt
	disease onset; C were more likely to be on antimalarials but less
	likely to be on corticasteroids. H, AA, and C used health care resources
	comparably. They had similar levels of pain and physical and mental
	functioning after adjusting for age, disease duration, income, education,
	social support, illness-related behaviors, and Systemic Lupus Activity
	Measure or SLAM scores. Conclusions: H and AA patients have more
	active SLE, at an earlier age of onset, and a less favorable socioeconomic-demographic
	structure (worse among the H than AA) which predispose them to a
	less favorable natural history.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Alarcon2000,
  author = {Alarcon, G. S.},
  title = {Unclassified or undifferentiated connective tissue disease},
  journal = {Best Practice \& Research in Clinical Rheumatology},
  year = {2000},
  volume = {14},
  pages = {125-137},
  number = {1},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Stallard2009,
  author = {Stallard, N.},
  title = {Simple tests for the external validation of mortality prediction
	scores},
  journal = {Statistics in Medicine},
  year = {2009},
  volume = {28},
  pages = {377-388},
  number = {3},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Sontheimer2005,
  author = {Sontheimer, R. D.},
  title = {Subacute cutaneous lupus erythematosus: 25-year evolution of a prototypic
	subset (subphenotype) of lupus erythematosus defined by characteristic
	cutaneous, pathological, immunological, and genetic findings},
  journal = {Autoimmunity Reviews},
  year = {2005},
  volume = {4},
  pages = {253-263},
  number = {5},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.10}
}

@ARTICLE{Hochberg1993,
  author = {Hochberg, Marc C.},
  title = {The history of lupus erythematosus},
  journal = {Lupus Foundation of America Newsletter},
  year = {1993},
  owner = {rebeccaspeckman},
  series = {Lupus Foundation of America Newsletter},
  timestamp = {2009.11.10}
}

@BOOK{Mooney1997,
  title = {Monte Carlo simulation},
  publisher = {Sage Publications},
  year = {1997},
  author = {Mooney, Christopher Z.},
  series = {Sage university papers series. Quantitative applications in the social
	sciences},
  address = {Thousand Oaks, Calif.},
  note = {96045873 Christopher Z. Mooney. ill. ; 22 cm. Includes bibliographical
	references (p. 99-110).},
  keywords = {Monte Carlo method.},
  owner = {rebeccaspeckman},
  timestamp = {2009.11.12}
}

@BOOK{Klippel2001,
  title = {Primer on the rheumatic diseases},
  publisher = {Arthritis Foundation},
  year = {2001},
  author = {Klippel, John H.},
  address = {Atlanta, GA},
  edition = {12th},
  note = {[edited by] John H. Klippel, MD ... [et al.]. cm.},
  owner = {rebeccaspeckman},
  timestamp = {2009.12.07}
}

@comment{jabref-meta: selector_journal:}

@comment{jabref-meta: selector_author:}

@comment{jabref-meta: selector_keywords:}

@comment{jabref-meta: selector_publisher:}

@comment{jabref-meta: groupsversion:3;}

@comment{jabref-meta: groupstree:
0 AllEntriesGroup:;
1 ExplicitGroup:Mixture modelling\;0\;Fraley1999\;Fraley2002\;McLachla
n2004\;Symons1981\;Tikka2007\;;
2 ExplicitGroup:Bernoulli mixture models\;0\;Carreira-Perpinan2000\;Gr
im2006\;Grim2007\;Juan2001\;Juan2004\;Juan2004a\;Seppanen2003\;;
2 ExplicitGroup:relevant?\;0\;Todem2007\;;
2 ExplicitGroup:MCMC LCA/MM\;0\;Garrett2000\;Zeger2003\;;
2 ExplicitGroup:conditional dependence models\;0\;;
2 ExplicitGroup:non-identifiability\;0\;DeBacker1999\;;
2 ExplicitGroup:local maxima\;0\;;
1 ExplicitGroup:Data generation\;0\;Cary\;Krummenauer1998\;Makoto/Taku
ji\;Qiu2006\;Savicky2006\;Scott1999\;Steinley2005\;Waller1999\;;
2 ExplicitGroup:Binary data generation\;0\;Chaganty2006\;Gange1995\;Jo
e2004\;Joe2006\;Leisch1998\;Lunn1998\;Park1996\;Snijders1991\;;
1 ExplicitGroup:Similarity measures\;0\;Baroni-Urbani1976\;Jackson1989
\;Real1996\;Su2001\;;
1 ExplicitGroup:Cluster analysis\;0\;Dubes1976\;Everitt1979\;He2005\;H
e2008\;;
2 ExplicitGroup:MC comparisons/studies\;0\;Bayne1980\;Dolcinar1998\;Do
lcinar1998a\;Milligan1981a\;Milligan1983\;;
2 ExplicitGroup:Binary data\;0\;;
2 ExplicitGroup:Reviews\;0\;Milligan1987\;Steinley2006a\;;
2 ExplicitGroup:Misc\;0\;Jain1987\;;
2 ExplicitGroup:Relation to prob methods\;0\;Banfield1993\;Bryant1991\
;Scott1971\;Steinley2006a\;Symons1981\;;
2 ExplicitGroup:Non-Kmeans-ish methods\;0\;Cesario2007\;Dutta2005\;Edw
ards1965\;Gowda1992\;Guha2000\;Orloci1967\;Shatovska2008\;Song2006\;Va
nMechelen2004\;Yeung2001a\;Yin2008\;;
2 ExplicitGroup:Alternative optimization criteria\;0\;Friedman1967\;Ma
ronna1974\;Marriott1971\;Marriott1982\;Scott1971\;Steinley2006a\;Windh
am1987\;;
2 ExplicitGroup:K-centroids\;0\;Baroni-Urbani1976\;Huang1998\;McRae197
1\;Steinley2003\;Steinley2004a\;Steinley2006a\;Steinley2007a\;Steinley
2008\;Steinley2008a\;Su2001\;;
2 ExplicitGroup:Fuzzy clustering\;0\;Manton2004\;;
2 ExplicitGroup:clustering tendency\;0\;Panayirci1983\;;
2 ExplicitGroup:Cluster validation\;0\;Bezdek1998\;Brusco2007a\;Datta2
003\;Dimitriadou2002\;Dubes1979\;Dubes1987\;Irigoien2008\;Jaksic1990\;
Lee1979\;;
3 ExplicitGroup:bootstrap, MC etc.\;0\;Bertrand2006\;Kuncheva2006\;Nem
ec1988\;Steinley2006\;Tibshirani2001\;Tonidandel2004\;;
3 ExplicitGroup:Null models\;0\;Gordon1994\;Hunter2004\;Jackson1992\;M
atthiessen2003\;Tibshirani2001\;Vassiliou1989\;;
3 ExplicitGroup:reviews (some have MC)\;0\;Halkidi\;Handl2005\;Milliga
n1981\;Milligan1985\;;
3 ExplicitGroup:cross-validation\;0\;Kapp2007\;Krieger1999\;;
3 ExplicitGroup:methods\;0\;Krzanowski1988\;Loganantharaj2006\;McShane
2002\;Steinley2007\;Strauss1982\;Yeung2001\;;
3 ExplicitGroup:ARI etc.\;0\;Brennan1974\;Fowlkes1983\;Hubert1985\;Kla
storin1985\;Mantel1967\;Morey1984\;Rand1971\;Steinley2004\;;
1 ExplicitGroup:Classification (to known classes)\;0\;Guvenir1998\;;
1 ExplicitGroup:Disease applications\;0\;Guvenir1998\;;
1 ExplicitGroup:Bayesian methods\;0\;Casella1992\;Chib1995\;;
1 ExplicitGroup:Neural networks\;0\;DeBacker1999\;;
1 ExplicitGroup:Numerical methods (?)\;0\;Dempster1977\;;
}