Can An ”Estimation Factor” Help Explain Cross-Sectional Returns?

We show in a theoretical model that the expected excess return on any asset depends on its covariance not only with the market portfolio, but also with changes in the representative agent’s estimate. In the empirical specification, this ”estimation factor” is based on realized growth in aggregate dividends and earnings. We test our model by using GMM and compare it to the Fama-French model. The results suggest that the estimation factor is priced. Moreover, the Hansen-Jagannathan distances show that the conditional and static versions of our derived model perform on a par with the corresponding versions of the Fama-French model.