Empirical Asset Pricing and Statistical Power in the Presence of Weak Risk Factors

102 Pages Posted: 25 Mar 2008

See all articles by A. Craig Burnside

A. Craig Burnside

Duke University - Department of Economics; University of Glasgow - Department of Economics; National Bureau of Economic Research (NBER)

Multiple version iconThere are 3 versions of this paper

Date Written: March 18, 2008

Abstract

Risk factors in many consumption-based asset pricing models display statistically weak correlation with the returns being priced. Some GMM procedures used to test these models have low power to reject misspecified stochastic discount factors (SDFs) when the covariance matrix of the asset returns with the risk factors has less than full column rank. Consequently, these estimators provide potentially misleading positive assessments of the SDFs. Two summary tests for failure of the rank condition have reasonable power, and lead to no Type I errors in Monte Carlo experiments.

Keywords: linear factor models, spurious factors, normalizations, consumption-based asset pricing, GMM

JEL Classification: C33, F31, G12

Suggested Citation

Burnside, Craig, Empirical Asset Pricing and Statistical Power in the Presence of Weak Risk Factors (March 18, 2008). Available at SSRN: https://ssrn.com/abstract=1107670 or http://dx.doi.org/10.2139/ssrn.1107670

Craig Burnside (Contact Author)

Duke University - Department of Economics ( email )

213 Social Sciences Building
Box 90097
Durham, NC 27708-0204
United States

University of Glasgow - Department of Economics

Adam Smith Building
Glasgow, Scotland G12 8RT
United Kingdom

National Bureau of Economic Research (NBER)

1050 Massachusetts Avenue
Cambridge, MA 02138
United States