Econometrics Modeling with Multilayer and Induced Network Data Restricted; Files Only

Abstract

Data abundance and complexity necessitate the identification of a system's operations mechanisms. In an economy, one of the most important driving forces of its mechanisms is related to economic entities interactions. The study of interactions can uncover multiple types of relationships and complicated patterns. Multilayer networks literature offers the mathematical and topological bases to represent interactions in network structures and evaluate their characteristics and diagnostics. On the other hand, Econometrics offers the flexibility to construct models able to incorporate networks and provide insights for causal inference.

Network connections can be utilized for inference when incorporated in economic models but they can also act as post-estimation instruments towards understanding estimated effects across entities and over time. Economic interactions are also dynamic and their evolution can be depicted in network formations. Network representations can reveal information such as the centrality of entities in the entire system, the entities' incoming and outgoing connections ratio, the distance and clustering among entities and the strength of the connections.

Economic crisis periods can be better diagnosed and possibly prevented with systemic risk monitoring. To evaluate risk transmission and study estimated spillover effects among economic entities, one can employ a network framework which can report on the economic situation. More specifically, on how shocks originating from the system's entities affect others and how these interactions have formulated the entire network.

The first chapter proposes the identification and estimation of social parameters, under a Multilayered Linear-in-Means model, when there are more than one types of social ties among economic agents. The second chapter replicates the code of an empirical paper and provides new findings using an extended version of the original dataset. The third chapter proposes the estimation of block specific spillover effects for credit risk monitoring.

Table of Contents

1 Estimating Social Effects in a Multilayered Linear-in-Means Model with Network Data . . 1

1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.2 Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

1.3 Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

1.4 Monte Carlo Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

1.5 Empirical Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

1.6 Appendix: Constrained Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

1.7 Appendix: Proofs of Main Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

1.8 Appendix: Auxiliary Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

1.9 Appendix: Monte Carlo Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

1.9.1 Supplementary Monte Carlo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

1.10 Appendix: Empirical Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

1.10 .1 Data Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

1.10.2 Supplementary Estimation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

2 Risk Transmission across European entities using Networks: A Wide Replication of Gross

and Siklos (2020) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

2.2 Replication Static results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

2.3 Replication Dynamic results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

2.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

3 Estimated spillover effects of block factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

3.2 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

3.3 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

3.3.1 The Dynamic Hierarchical Factor Model (DHFM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

3.3.2 A Vector Autoregressive (VAR) model with block factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96

3.4 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

3.4.1 Full-sample Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

3.4.2 Sub-sample Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122

3.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144

3.6 Appendix: Assumed zero correlations under the DHFM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145

About this Dissertation

Rights statement

• Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.

School	Laney Graduate School
Department	Economics
Degree	Ph.D.
Submission	Dissertation
Language	English
Research Field	Economics, General
Palabra Clave	social effects spillover effects networks contagion multilayer
Committee Chair / Thesis Advisor	Elena Pesavento, Emory University David T. Jacho-Chavez, Emory University
Committee Members	Weihua An, Emory University Kim P. Huynh, Director, Bank of Canada

Última modificación

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	File download under embargo until 22 May 2025	2023-04-06 11:51:36 -0400	File download under embargo until 22 May 2025

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions