Predictive regressions under asymmetric loss: factor augmentation and model selection

The paper discusses the specifics of forecasting with factor-augmented predictive regressions under general loss functions. In line with the literature, we employ principal component analysis to extract factors from the set of predictors. We additionally extract information on the volatility of the series to be predicted, since volatility is forecast-relevant under non-quadratic loss functions. To ensure asymptotic unbiasedness of forecasts under the relevant loss, we estimate the predictive regression by minimizing the in-sample average loss. Finally, to select the most promising predictors for the series to be forecast, we employ an information criterion tailored to the relevant loss. Using a large monthly data set for the US economy, we assess the proposed adjustments in a pseudo out-of-sample forecasting exercise for various variables. As expected, the use of estimation under the relevant loss is effective. Using an additional volatility proxy as predictor and conducting model selection tailored to the relevant loss function enhances forecast performance significantly.