Do Cross-Sectional Predictors Contain Systematic Information?

Abstract

Firm-level variables that predict cross-sectional stock returns, such as price-to-earnings and short interest, are often averaged and used to predict market returns. Using various samples of cross-sectional predictors and accounting for the number of predictors and their interdependence, we find only weak evidence that cross-sectional predictors make good time-series predictors, especially out-of-sample. The results suggest that cross-sectional predictors do not generally contain systematic information.

Keywords: Return predictability, data snooping, statistical bias, market risk premium.

JEL Classification: G00, G14, L3, C1