Estimating Future Performance: The Shrinkage-Adjusted Sharpe Ratio

Estimation error is a central problem in mutual fund selection: past return parameters are very noisy estimates of the corresponding out-of-sample parameters. Fortunately, statistical “shrinkage” can improve estimation. The shrinkage-adjusted Sharpe ratio (SAS) is based on the prescription that shrinkage should usually be applied to the gross sample returns, but not to fees, which are typically known. The SAS significantly improves out-of-sample performance relative to existing methods.