Application of Dynamic Factor Models to Macroeconomics Restricted; Files Only

Abstract

This dissertation consists of three chapters on dynamic factor models and macroeconomics. The first chapter investigates the effects of fiscal policy on asset prices using structural dynamic factor model (SDFM) with cointegrated factors I identify fiscal shock using narrative approach; specifically, I use military spending (war dates) as a fiscal policy variable. The results of the IRFs show that both stock prices and house prices responded positively to fiscal shock and the effects were persistent. The result implies that fiscal pol- icy leads to a boom in housing and stock markets. This paper highlighted that data-rich models play an important role in obtaining true IRFs as they provide a more accurate representation of economic concepts.

In the second chapter, I use a time-varying dynamic factor model with stochastic volatility, which allows us to measure the relative contribution of national, regional, and idiosyncratic state-specific factors to the variations in state-level US house prices and how this variation has evolved over time using quarterly state-level and metropolitan statistical areas (MSAs) data from 1975 to 2021 and applying the Gibbs sample algorithm via Bayesian Markov Chain Monte Carlo (MCMC) simulations. I find that the U.S housing market is a national phenomenon as national factor contributes to large house price movements via variance decomposition in full sample and sub-samples. This shows that national factors play an important role in explaining the volatility in US state-level and MSA-level housing prices.

In the third chapter, I estimate the spillover effects among European sovereigns, financial, non-financial institutions when extracting global and block-specific common factors.

By allowing for the presence of global and block common factors, we can evaluate spillovers when shocks are originating from global common factors, block common factors and the idiosyncratic component. In addition, this way we can dive more into the purely idiosyncratic spillover effects which are free of block-related spillovers. Our analysis allows us to visualize our findings using heatmaps and compare spillover effects during two periods, the European financial and sovereign crisis period as well as the Covid-19 period.

Table of Contents

1 Fiscal Policy and Asset Prices in a Dynamic Factor Model with Cointegrated Factors 1

1.1 Introduction .......................................................... 1

1.2 Literature Review ..................................................... 4

1.3 Model Specification ................................................... 6

1.3.1 Estimation ...................................................... 8

1.3.2 Factor Normalization and Estimation ............................. 8

1.3.3 Estimating the Number of Factors and Dynamic Shocks ............. 9

1.3.4 Estimation of Impulse Response Functions ....................... 10

1.3.5 Identification of Government Spending Shocks ................... 11

1.4 Data ................................................................ 13

1.5 Results .............................................................. 13

1.5.1 Number of Factors .............................................. 14

1.5.2 Impulse Response of Fiscal Policy Shock ........................ 15

1.5.3 Variance Decomposition Analysis ................................ 16

1.5.4 Robustness Check using Full Sample Data (1959-2021) ............ 17

1.5.5 Impulse Response Function using a SVAR Model (Comparison) . . 18

1.6 Conclusion ........................................................... 21

2 Comovement and Transmission of Shocks in US Housing Market 22

2.1 Introduction ......................................................... 22

2.2 Model Specification .................................................... 26

2.2.1 Model Estimation .................................................. 27

2.2.2 Variance decomposition ............................................ 28

2.2.3 Data .............................................................. 29

2.3 Results .................................................................. 30

2.3.1 Variance Decomposition ............................................ 30

2.3.2 Comovement of State-Level House Price and the National Factor . . 39

2.3.3 Posterior Volatilities of National and Regional Factors ........... 42

2.3.4 Variance Decomposition of Metropolitan House Prices ............... 43

2.3.5 Posterior Volatilities of National and Regional Factors (MSA House Prices) ................................................................ 45

2.4 Conclusion ............................................................... 46

3 Extracting global and block common factors to monitor the connectedness of credit risk in Europe 47

3.1 Introduction ............................................................. 47

3.2 Literature Review ............................................... 50

3.3 Measures of Connectedness ................................................ 51

3.3.1 Connectedness in a Static form of a Dynamic Factor Model .......... 51

3.3.2 Connectedness in a Static form of a Dynamic Factor Model with Blocks ........................................ 54

3.4 Empirical Results ........................................................ 56

3.4.1 Variance decomposition of a shock in the global factor . 57

3.4.2 Variance decomposition of a shock in the idiosyncratic components 59

3.4.3 Variance decomposition of a shock in the block factors . 64

3.4.4 Variance decomposition of a shock to the global factor with block

factors (Financial Crisis vs COVID Period) ...................... 67

3.5 Conclusions .............................................................. 73

Appendix A Data Description and Methods 74

Appendix B Results of 3-block factor model 78

Bibliography 85

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
Stichwort	House Prices Asset Prices Connectedness Macroeconomics Time-Varying Dynamic Factor Model Credit Default Spread Stochastic Volatility Fiscal Policy
Committee Chair / Thesis Advisor	Juan Rubio Ramirez, Emory University Elena Pesavento, Emory University
Committee Members	Kaiji Chen, Emory University

Zuletzt geändert

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	File download under embargo until 22 May 2026	2024-04-22 15:33:23 -0400	File download under embargo until 22 May 2026

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions