Variable Selection in Cross‐Section Regressions: Comparisons and Extensions

33 Pages Posted: 28 Oct 2014

See all articles by Thomas Deckers

Thomas Deckers

University of Bonn

Christoph Hanck

University of Dortmund - Department of Statistics

Date Written: December 2014

Abstract

Cross‐section regressions often examine many candidate regressors. We use multiple testing procedures (MTPs) controlling the false discovery rate (FDR) — the expected ratio of false to all rejections — so as not to erroneously select variables because many tests were performed, yielding a simple model selection procedure. Simulations comparing the MTPs with other common model selection criteria demonstrate that, for conventional tuning parameters of the selection procedures, only MTPs consistently control the FDR, but have slightly lower power. In an empirical application to growth, MTPs and PcGets/Autometrics identify similar growth determinants, which differ somewhat from those obtained by Bayesian Model Averaging.

Suggested Citation

Deckers, Thomas and Hanck, Christoph, Variable Selection in Cross‐Section Regressions: Comparisons and Extensions (December 2014). Oxford Bulletin of Economics and Statistics, Vol. 76, Issue 6, pp. 841-873, 2014, Available at SSRN: https://ssrn.com/abstract=2515613 or http://dx.doi.org/10.1111/obes.12048

Thomas Deckers (Contact Author)

University of Bonn ( email )

Regina-Pacis-Weg 3
Postfach 2220
Bonn, D-53012
Germany

Christoph Hanck

University of Dortmund - Department of Statistics ( email )

D-44221 Dortmund
Germany

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
1
Abstract Views
248
PlumX Metrics