Bayesian exploratory factor analysis

Conti, Gabriella and Frühwirth-Schnatter, Sylvia and Heckman, James J. and Piatek, Rémi (2014) Bayesian exploratory factor analysis. Journal of Econometrics, 183 (1). pp. 31-57. ISSN 03044076

Preview

PDF Conti_etal_2014_JE.pdf Download (732kB)

Abstract

This paper develops and applies a Bayesian approach to Exploratory Factor Analysis that improves on ad hoc classical approaches. Our framework relies on dedicated factor models and simultaneously determines the number of factors, the allocation of each measurement to a unique factor, and the corresponding factor loadings. Classical identification criteria are applied and integrated into our Bayesian procedure to generate models that are stable and clearly interpretable. A Monte Carlo study confirms the validity of the approach. The method is used to produce interpretable low dimensional aggregates from a high dimensional set of psychological measurements. (authors' abstract)

Item Type:	Article
Additional Information:	This paper is forthcoming in Journal of Econometrics. We thank the editor, John Geweke, and two anonymous referees for comments. This work was presented at the third European Seminar on Bayesian Econometrics (November 2012, Vienna). We thank the participants for their helpful comments, and especially Xiao-Li Meng and our discussant Jesús Crespo Cuaresma. We also gratefully acknowledge support from NIH R01 HD054702 and R37 HD065072, the American Bar Foundation, The J.B. and M.K. Pritzker Foundation, the Geary Institute at University College Dublin, a grant from the European Research Council DEVHEALTH-269874, and an anonymous funder. The research of the second author was partly funded by the Austrian Science Fund (FWF): S10309-G16. The views expressed in this paper are those of the authors and not necessarily those of the funders. A Web Appendix containing additional material is available at http://heckman.uchicago.edu/BayesFA
Keywords:	Bayesian Factor Models / Exploratory Factor Analysis / Identifiability / Marginal Data Augmentation / Model Expansion / Model Selection
Classification Codes:	JEL C11, C38, C63
Divisions:	Departments > Finance, Accounting and Statistics > Statistics and Mathematics
Version of the Document:	Accepted for Publication
Variance from Published Version:	Minor
Depositing User:	Elena Simukovic
Date Deposited:	05 May 2015 07:38
Last Modified:	15 Sep 2017 06:55
Related URLs:	Publisher
FIDES Link:	https://bach.wu.ac.at/d/research/results/70646/
URI:	https://epub.wu.ac.at/id/eprint/4521

Actions

View Item

Downloads