Notes: This table reports predictive $R2$⁠, their standard deviation and statistical significance, and the annualized Sharpe ratios (SR) implied by Martin and Wagner’s (2019) and Kadan and Tang’s (2020) theory-based approaches, the five machine learning models, and a hybrid approach in which the theory-based RPE serve as additional features in the machine learning models (ML with theory features). The standard deviation of the $Roos,s2×100$ (Std Dev) is calculated based on the annual test samples. The SR refer to a zero-investment strategy long in the portfolio of stocks with the highest excess return prediction and short in the portfolio of stocks with the lowest excess return prediction. The p-values are associated with a test of the null hypothesis that the respective excess return prediction has no explanatory power over the zero forecast, $E(Roos,s2)≤0$⁠. The RPE refer to a one-month investment horizon and are computed at the monthly (EOM) frequency. The out-of-sample testing period starts in January 1998 and ends in November 2018. The features are rank-scaled as described in Section 3.6. The machine learning results are obtained using the short training scheme depicted in Figure 3.