Optimal Portfolio Choice with Estimation Risk: No Risk-free Asset Case
52 Pages Posted: 10 Feb 2016 Last revised: 29 Oct 2018
Date Written: October 27, 2018
For the popular mean-variance portfolio choice problem in the case without a risk-free asset, we develop a new optimal portfolio rule that is designed to mitigate estimation risk. We compare its out-of-sample performance, both theoretically and empirically, with that of other portfolio rules. In both calibrations and real datasets, we show that our new rule performs well relative to others. Our results suggest the importance of explicitly taking into account the impact of estimation risk when forming an optimal portfolio. In addition, we derive the exact distribution of the out-of-sample returns and provide the explicit expression of the out-of-sample expected utility of various optimal portfolio rules, which offers analytical insights into portfolio construction and performance evaluation.
Keywords: portfolio choice, estimation risk, global minimum-variance portfolio, 1/N rule
JEL Classification: G11, G12, C11
Suggested Citation: Suggested Citation