| Abstract: |
In this paper, we study the risk-return relationship in monthly U.S. stock
returns (1928:1— 2004:12) using GARCH-in-Mean models. In particular, we
consider the robustness of the relationship with respect to the omission of
the intercept term in the equation for the expected excess return recently
recommended by Lanne and Saikkonen (2006). The existence of the relationship
is quite robust, but its estimated strength is dependent on the prior belief
concerning the intercept. This is the case in particular in the first half of
the sample period, where also the coefficient of the relative risk aversion is
found to be smaller and the equity premium greater than in the latter half. |