Robust Predictive Regression

A large literature studies the predictability of stock returns by other lagged nancialvariables in a predictive regression setting. A common feature of widely used testingprocedures is a failing robustness, which may lead to misleading conclusions determinedby the particular features of a small subfraction of the data. We propose a new generalmethod to deal with this problem based on the robust subsampling approach. The methodimplies robust condence intervals and inference results. It is applicable both in the multi-predictor context and in settings with nearly integrated regressors. Simulation evidenceconrms the higher accuracy and efficiency of our robust testing approach for typicalapplications in which the data may follow only approximately a predictive regressionmodel.[...]