Fixed Effects Vector Decomposition: Reply

37 Pages Posted: 9 Jan 2011 Last revised: 11 Jan 2011

See all articles by Thomas Pluemper

Thomas Pluemper

Vienna University of Economics and Business - Department of Socioeconomics; University of Essex - Department of Government

Vera E. Troeger

University of Essex - Department of Government

Date Written: January 9, 2011

Abstract

This article reinforces our 2007 Political Analysis publication in demonstrating that the fixed effects vector decomposition (FEVD) procedure outperforms any other estimator in estimating models which suffer from the simultaneous presence of time-varying variables correlated with unobserved unit effects and time-invariant variables. We compare the finite sample properties of FEVD not only to the Hausman-Taylor (HT) estimator but also to the pre-test estimator and the shrinkage estimator suggested by Breusch et al. (BWNK) in this symposium. Moreover, we correct Greene’s and BWNK’s discussion of FEVD’s asymptotic and finite sample properties.

Keywords: Fixed Effects Vector Decomposition, Efficiency, Unbiasedness, Consistency, Instruments

JEL Classification: C23, C15

Suggested Citation

Plümper, Thomas and Troeger, Vera E., Fixed Effects Vector Decomposition: Reply (January 9, 2011). Available at SSRN: https://ssrn.com/abstract=1737480 or http://dx.doi.org/10.2139/ssrn.1737480

Thomas Plümper (Contact Author)

Vienna University of Economics and Business - Department of Socioeconomics ( email )

Vienna
Austria

University of Essex - Department of Government ( email )

Wivenhoe Park
Colchester, Essex CO4 3SQ
United Kingdom

HOME PAGE: http://www.polsci.org/pluemper

Vera E. Troeger

University of Essex - Department of Government ( email )

Wivenhoe Park
Colchester CO4 3SQ, CO4 3SQ
United Kingdom

Register to save articles to
your library

Register

Paper statistics

Downloads
419
Abstract Views
1,810
rank
69,349
PlumX Metrics