Forecasting Stock Returns with Model Uncertainty and Parameter Instability
35 Pages Posted: 21 Sep 2017 Last revised: 8 Dec 2019
Date Written: September 16, 2019
Abstract
We compare several representative sophisticated model averaging and variable selection techniques of forecasting stock returns. When estimated traditionally, our results confirm that the simple combination of individual predictors is superior. However, sophisticated models improve dramatically once we combine them with the historical average and take parameter instability into account. An equal weighted combination of the historical average with the standard multivariate predictive regression estimated using the average windows method, for example, achieves a statistically significant monthly out-of-sample R2 of 1.10% and annual utility gains of 2.34%. We obtain similar gains for predicting future macro economic conditions.
Keywords: Market Efficiency, Asset Pricing, Equity Premia Predictability, Forecast Combination, Model Uncertainty, Parameter Instability, WALS
JEL Classification: G17, G12, G02, C58
Suggested Citation: Suggested Citation