According to the CAPM, risk is measured by the beta, and the relation between required expected return and beta is linear. This paper examines the conditional relationship between beta and return in the US stock market. The conditional covariances and variances used to estimate beta are modeled as an ARCH process. The beta return relationship is tested upon the sign of the excess market return. The implication of the sign of the excess market return follows Morelli (2011). This study shows the importance of recognizing the sign of the excess market return when testing the beta-return relationship. The approach also allows us to distinguish the size effect and the effect of economic cycles.
Category - Bing XIAO
Université d’Auvergne, France