Predictive Regressions with Time-Varying Coefficients

Abstract

We evaluate predictive regressions that explicitly consider the time-variation of coefficients in a comprehensive Bayesian framework. This allows for fast and consistent adjustment of regression coefficients to changes in the underlying economic relationships. For monthly returns of the S&P 500 index, we demonstrate statistical as well as economic evidence of out-of-sample predictability: relative to an investor using the historic mean an investor using our methodology could have earned consistently positive utility gains (between 1.8 and 5.8% p.a. over different time periods). Furthermore, we show that predictive models with constant coefficients are dominated by models with time-varying coefficients.

Keywords: Empirical asset pricing, equity return prediction, Bayesian econometrics

JEL Classification: G12, C11