Endogenously Clustered Factor Approach to Macroeconomics

Public Deposited
Analytics

Downloadable Content

Download PDF
Request Version for Screen Reader
Last Modified
  • March 20, 2019
Creator
  • Savascin, Ozge
    • Affiliation: College of Arts and Sciences, Department of Economics
Abstract
  • This dissertation constructs a novel factor approach to study the comovements of macroeconomic variables and introduces its two practical applications. Factor models have become useful tools for studying international business cycles. Block factor models can be especially useful as the zero restrictions on the loadings of some factors may provide some economic interpretation of the factors. These models, however, require the econometrician to predefine the blocks, leading to potential misspecification. In my coauthored paper, we propose an alternative model in which the blocks are chosen endogenously. The model is estimated in a Bayesian framework using a hierarchical prior which allows series-level covariates to influence and explain how the series are grouped. Using similar international business cycle data as Kose, Otrok, and Whiteman (2005) we find our country clusters differ in important ways to those identified by geography alone. In particular, we find that similarities in institutions -- e.g., legal systems, language diversity -- may be just as important as physical proximity for analyzing business cycle comovements. In another application, I use the endogenously clustered dynamic factor model to gain a better understanding of commodity price comovements and their determinants. From a large dataset of commodity prices I extract the fundamental sources behind the price dynamics and find that commodity price comovements are mostly the result of sparse cluster factors that represent correlations of distinct group of commodities. Endogenous clustering of these groups does not represent the standard narrow classifications (indexes) of commodity prices as defined by statistical agencies (e.g. International Financial Statistics, Bureau of Labor Statistics). Characterization analysis on these factors identifies a wide range of macroeconomic variables like crude oil prices, fertilizer prices, and the federal funds rate as possible sources of commodity price comovements.
Date of publication
Keyword
DOI
Resource type
Rights statement
  • In Copyright
Advisor
  • Owyang, Michael
  • Froyen, Richard T.
  • Hill, Jonathan
  • Francis, Neville
  • Ghysels, Eric
Degree
  • Doctor of Philosophy
Degree granting institution
  • University of North Carolina at Chapel Hill Graduate School
Graduation year
  • 2012
Language
Access right
  • This item is restricted from public view for 1 year after publication.
Date uploaded
  • April 26, 2017

Relations

Parents:

This work has no parents.

Items

Thumbnail Title Date Uploaded Visibility Actions
file details Savascin_unc_0153D_12954.pdf 2019-04-10 Public Download