Enhanced Mean–Variance Portfolios:
A Controlled Integration of Quantitative Predictors

Abstract

The intuitiveness and practicality of mean–variance portfolios largely depend on the accuracy of moment estimates, which are subject to large estimation errors and are conditional on time. The authors propose a model that accounts for factor dynamics in a Bayesian setting, in which they endogenously derive the effect of estimation accuracy on the posterior distribution from a linear predictive regression model. By doing so, they capture upside return potential for periods of high factor-explained variance, while constraining downside risk for periods of low predictive quality. Results are robust in a simulation and an empirical setting.