Evaluating Hedge Funds with Pooled Benchmarks

The evaluation of hedge fund performance is challenging given the flexible nature of hedge funds’ strategies and their lack of operational transparency. The conventional method relies on the use of a single benchmark model that is typically identified using model selection criteria. This approach is problematic since inference about skill is inevitably contaminated by the error in the benchmark model. To address this concern, we propose a model pooling approach. The paper employs a fund-specific benchmark obtained by pooling a set of diverse models to evaluate hedge fund performance. The weights assigned to the individual models in the pool are based on the log score criterion, a measure of the conditional performance of a model. We illustrate the advantages of a pooled benchmark over alternative approaches, including the widely used Fung and Hsieh (2004) model and stepwise regression methods, in a variety of contexts including a real-time investment strategy, hedge fund replication, and fund failure prediction.