Can Omitted Risk Factors Explain the January Effect? A Stochastic Dominance Approach

Abstract

This paper provides a direct test of the hypothesis that large January returns can be attributed to omitted risk factors. Data from 1926–1991 show that the January return in the smallest decile of NYSE firms dominates the January returns for all other deciles by the first-order stochastic dominance. Similarly, January returns in all deciles (with the exception of ninth and tenth deciles) dominate non-January returns by first-, second-, or third-order stochastic dominance. The presence of stochastic dominance by January returns suggests that the omitted risk factors are not likely to explain the January effect.