The Shrinkage Adjusted Sharpe Ratio: An Improved Method for Mutual Fund Selection

Abstract

Mutual fund selection is a notoriously difficult task, because past performance is a poor predictor of future performance. We propose a fund performance measure that incorporates a simple idea: shrinkage, in the sense of Bayes-James-Stein, should be applied to gross return parameters, but not to fees, which are known. The proposed Shrinkage Adjusted Sharpe ratio (SAS) substantially improves the prediction of out-of-sample performance relative to existing methods. The best prediction is obtained when fees are weighed five times heavier than sample returns.