Kosowski, Robert; Timmermann, Allan; Wermers, Russ; White, Hal (2005) examine whether mutual fund "stars" can truly pick stocks using a bootstrap analysis of U.S. open-end, domestic-equity mutual funds from 1975 to 2002. They find that a significant minority of fund managers possess genuine stock-picking skills, as their high alphas persist even after controlling for luck. The bootstrap method is used to assess the significance of fund alphas, accounting for non-normal distributions and heteroskedasticity. The results show that top-performing funds have alphas that are unlikely to be due to chance alone. The study also finds that high-alpha funds are more likely to be skill-based, while low-alpha funds are often due to cost inefficiencies. The analysis highlights the importance of controlling for luck in performance evaluation and shows that some fund managers have superior stock-picking abilities. The results are robust across different models and bootstrap methods, indicating that the superior performance of certain funds is not merely due to luck. The study concludes that the bootstrap approach is crucial for accurately assessing fund performance and identifying true skill in stock selection.Kosowski, Robert; Timmermann, Allan; Wermers, Russ; White, Hal (2005) examine whether mutual fund "stars" can truly pick stocks using a bootstrap analysis of U.S. open-end, domestic-equity mutual funds from 1975 to 2002. They find that a significant minority of fund managers possess genuine stock-picking skills, as their high alphas persist even after controlling for luck. The bootstrap method is used to assess the significance of fund alphas, accounting for non-normal distributions and heteroskedasticity. The results show that top-performing funds have alphas that are unlikely to be due to chance alone. The study also finds that high-alpha funds are more likely to be skill-based, while low-alpha funds are often due to cost inefficiencies. The analysis highlights the importance of controlling for luck in performance evaluation and shows that some fund managers have superior stock-picking abilities. The results are robust across different models and bootstrap methods, indicating that the superior performance of certain funds is not merely due to luck. The study concludes that the bootstrap approach is crucial for accurately assessing fund performance and identifying true skill in stock selection.