On the Economic Significance of Stock Return Predictability

Abstract We study the effects of time-varying volatility and investment horizon on economic significance stock market return predictability from perspective Bayesian investors. Using a vector autoregression framework with stochastic (SV) in returns predictor variables, we assess broad set twenty-six predictors both in-sample out-of-sample designs. Volatility are critically important for assessing predictors, as these factors affect how an investor learns about she chooses to invest based forecasts. find that statistically strong can be economically unimportant if they tend take extreme values high periods, have low persistence, or follow distributions fat tails. Several popular exhibit properties such their impressive statistical results do not translate into large gains. also demonstrate incorporating SV leads substantial utility gains real-time forecasting.