A Comparison of Two Averaging Techniques with an Application to Growth Empirics

CentER Discussion Paper Series No. 2008-39

42 Pages Posted: 17 Apr 2008  

J.R. Magnus

VU University Amsterdam - Faculty of Economics and Business Administration

Owen Powell

University of Vienna - Department of Economics; Vienna Center for Experimental Economics

Patricia Prufer

CentERdata; Tilburg University

Date Written: April 3, 2008

Abstract

Empirical growth research faces a high degree of model uncertainty. Apart from the neoclassical growth model, many new (endogenous) growth models have been proposed. This causes a lack of robustness of the parameter estimates and makes the determination of the key determinants of growth hazardous. The current paper deals with the fundamental issue of parameter estimation under model uncertainty, and compares the performance of various model averaging techniques. In particular, it contrasts Bayesian model averaging (BMA) - currently one of the standard methods used in growth empirics - with weighted-average least squares (WALS), a method that has not previously been applied in this context.

Keywords: model averaging, Bayesian analysis, growth determinants

JEL Classification: C51, C52, C13, C11

Suggested Citation

Magnus, J.R. and Powell, Owen and Prufer, Patricia, A Comparison of Two Averaging Techniques with an Application to Growth Empirics (April 3, 2008). Available at SSRN: https://ssrn.com/abstract=1121642 or http://dx.doi.org/10.2139/ssrn.1121642

Jan R. Magnus (Contact Author)

VU University Amsterdam - Faculty of Economics and Business Administration ( email )

De Boelelaan 1105
Amsterdam, 1081HV
Netherlands

Owen Powell

University of Vienna - Department of Economics ( email )

Bruennerstrasse 72
Vienna, A-1210
Austria

Vienna Center for Experimental Economics ( email )

Oskar-Morgenstern-Platz 1
Vienna, Vienna 1090
Austria

Patricia Prufer

Tilburg University ( email )

Department of Economics
CentER
Tilburg, 5032 RE
Netherlands

HOME PAGE: http://center.uvt.nl/phd_stud/prufer/

CentERdata ( email )

PO Box 90153
Tilburg, NL 5000 LE
Netherlands

Register to support our free research

Register

Paper statistics

Downloads
93
rank
250,098
Abstract Views
569
PlumX