“evaluating and Predicting the Failure Probabilities of Hedge Funds”

Hedge funds have the most sophisticated risk management practices; however, hedge funds also appear to have a short lifetime relative to other managed funds. In this study, we investigate the failure probabilities of hedge funds—particularly the failures due to financial distress. We forecast the failure probabilities of hedge funds using both a proportional hazard model and a logistic model. By utilizing a signal detection model and a relative operating characteristic curve as the prediction accuracy metrics, we found that both of the models have predictive power in the out-of-sample test. The proportional hazard model, in particular, has stronger predictive power, on average. JEL Classification: G33, G14, G17