RT Journal Article
SR Electronic
T1 Geometric Mean Maximization: A Note on Expected, Observed, and Simulated Performance
JF The Journal of Investing
FD Institutional Investor Journals
SP 87
OP 94
DO 10.3905/joi.2021.1.175
VO 30
IS 4
A1 Johnston, Ken
A1 Hatem, John
YR 2021
UL http://joi.pm-research.com/content/30/4/87.abstract
AB This article discusses why the maximization of any portfolio optimization model cannot be isolated from the investor’s risk perception. To allow for differing risk preferences among investors, the efficient frontiers of Sharpe ratio maximization (SRM) and geometric mean maximization (GMM) are the appropriate metrics for making a comparison. The authors demonstrate that, for a given level of risk, the two optimization techniques will choose the same portfolio asset weights. Although GMM provides investors with a different way to approach portfolio optimization (maximizing terminal wealth), it is not a competing portfolio optimization technique to mean–variance maximization but, rather, a complementary one.TOPICS: Portfolio construction, quantitative methods, statistical methods, performance measurementKey Findings▪ Comparison of portfolio optimization techniques require a predetermined risk level.▪ Geometric mean and Sharpe ratio maximization lead to the same portfolio weights for a given level of risk.▪ Geometric mean maximization expands the opportunity set of the mean–variance space.