On the Effect of Prior Assumptions in Bayesian Model Averaging with Applications to Growth Regression

25 Pages Posted: 20 Apr 2016

Multiple version iconThere are 2 versions of this paper

Date Written: June 1, 2007

Abstract

This paper examines the problem of variable selection in linear regression models. Bayesian model averaging has become an important tool in empirical settings with large numbers of potential regressors and relatively limited numbers of observations. The paper analyzes the effect of a variety of prior assumptions on the inference concerning model size, posterior inclusion probabilities of regressors, and predictive performance. The analysis illustrates these issues in the context of cross-country growth regressions using three datasets with 41 to 67 potential drivers of growth and 72 to 93 observations. The results favor particular prior structures for use in this and related contexts.

Keywords: Educational Technology and Distance Education, Geographical Information Systems, Statistical & Mathematical Sciences, Science Education, Scientific Research & Science Parks

Suggested Citation

Ley, Eduardo and Steel, Mark F.J., On the Effect of Prior Assumptions in Bayesian Model Averaging with Applications to Growth Regression (June 1, 2007). World Bank Policy Research Working Paper No. 4238, Available at SSRN: https://ssrn.com/abstract=991430

Eduardo Ley (Contact Author)

World Bank ( email )

1818 H Street, NW
Washington, DC 20433
United States

HOME PAGE: http://eWorldNet.org

Mark F.J. Steel

University of Warwick ( email )

Coventry CV4 7AL
United Kingdom