Research Papers
"Maximum Drawdown as Predictor of Mutual Fund Performance and Flows"
Co-authored with Tim Riley
Financial Analysts Journal, 2022
Abstract: Mutual funds’ maximum drawdowns are persistent, indicative of manager skill, and predictive of subsequent performance. Among funds with relatively strong past performance, those with relatively low past maximum drawdowns, on average, have an out‑of‑sample alpha of 2.40% per year. That alpha is magnified when markets are turbulent—a time during which manager skill should be most valuable. Investors are averse to drawdown risk. After controlling for typical measures of past performance, fund flows remain a decreasing function of maximum drawdowns, particularly among investors with greater risk aversion and during times of heightened risk aversion.
"Hedge Fund Performance Prediction with Machine Learning"
Co-authored with Alexey Malakhov and Tim Riley
Abstract: We use machine learning to dynamically identify and optimally combine the predictors of hedge fund performance. The portfolio formed based on the machine learning models has an out‑of‑sample alpha of 7.8% per year. The importance of each predictor varies over time, but among the 22 predictors we consider, the consistently important predictors are average return, maximum return, alpha, systematic risk, and beta activity. Machine learning provides valuable, unique information about future hedge fund performance that is not captured by individual predictors.
"Do Hedge Funds Bet Against Beta?"
Co-authored with Alexey Malakhov and Tim Riley
International Review of Economics and Finance, 2024
Abstract: Differences in conditions within the mutual fund and hedge fund industries should lead to different approaches with respect to the low beta anomaly. We find that, unlike most mutual funds, the average hedge fund tends to benefit considerably from the anomaly. About 2.3% per year of apparent alpha for the average hedge fund can be attributed to the low beta anomaly rather than manager skill. Low skill managers rely the most on the anomaly to generate returns, with the most reliant underperforming the least reliant by 5.9% per year.
