Asset Returns Under Model Uncertainty: Evidence from the Euro Area, the US and the UK

We analyze predictability of risk premium in the context of model uncertainty. Using data for the euro area, the US and the UK, we show that there is a large amount of model uncertainty and one can improve the forecasts of stock returns with a Bayesian Model Averaging (BMA) approach. The empirical evidence for the euro area suggests that several macroeconomic, financial and macro-financial variables are consistently among the most prominent determinants of risk premium. As for the US, only a few number of predictors play an important role. In the case of the UK, future stock returns are better forecasted by financial variables. These results are corroborated for both the M-open and the M-closed perspectives, different model priors and in the context of “in-sample” and “out-of-sample” forecasting. Finally, we highlight that the predictive ability of the BMA framework is stronger at longer periods, and clearly outperforms the constant expected returns and the autoregressive benchmark models.