Mutual Funds’ Conditional Performance Free of Data Snooping Bias
107 Pages Posted: 11 Jan 2021 Last revised: 15 Feb 2024
Date Written: November 25, 2020
Abstract
We introduce a test to assess mutual funds’ “conditional” performance that is based on updated information and corrects data snooping bias. Our method, named the functional False Discovery Rate “plus” (fFDR+), incorporates fund characteristics in estimating fund performance free of data snooping bias. Simulations suggest that the fFDR+ controls well the ratio of false discoveries and gains considerable power over prior methods that do not account for extra information. Portfolios of funds selected by the fFDR+ outperform other tests not accounting for information updating, highlighting the importance of evaluating mutual funds from a conditional perspective.
Keywords: Multiple testing, Functional false discovery rate, Informative covariates, Mutual funds, Alphas
JEL Classification: C11, C12, G23
Suggested Citation: Suggested Citation