Robust Econometric Inference for Stock Return Predictability

This study examines stock return predictability via lagged financial variables with unknown stochastic properties. We conduct a battery of predictability tests for US stock returns during the 1927-2012 period, proposing a novel testing procedure which: i) robustifies inference to the degree of persistence of the employed regressors, ii) accommodates testing the joint predictive ability of financial variables in multiple regression, iii) is easy to implement as it is based on a linear estimation procedure and iv) can be also used for long-horizon predictability tests. We provide some evidence in favor of short-horizon predictability in the full sample period. Nevertheless, this evidence almost entirely disappears in the post-1952 period. Moreover, predictability becomes weaker, not stronger, as the predictive horizon increases