Working Paper: CEPR ID: DP7780
Authors: Andrew J. Patton; Tarun Ramadorai
Abstract: We propose a new method to capture changes in hedge funds' exposures to risk factors, exploiting information from relatively high frequency conditioning variables. Using a consolidated database of nearly 15,000 individual hedge funds between 1994 and 2009, we find substantial evidence that hedge fund risk exposures vary significantly across months. Our new method also reveals that hedge fund risk exposures vary within months, and capturing this variation significantly improves the fit of the model. The proposed method outperforms an optimal changepoint approach to capturing time-varying risk exposures, and we find evidence that there are gains from combining the two approaches. We find that the cost of leverage, the carry trade return and the recent performance of equity indices are the most important drivers of changes in hedge fund risk exposures.
Keywords: Beta; Performance Evaluation; Structural Breaks; Time-Varying Risk
JEL Codes: C22; G11; G23
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| hedge fund risk exposures vary across months (G23) | hedge fund risk exposures vary significantly (G23) |
| hedge fund risk exposures vary within months (G23) | hedge fund risk exposures vary significantly (G23) |
| cost of leverage, carry trade returns, and equity index performance (G12) | changes in hedge fund risk exposures (G11) |
| modeling technique (C50) | accuracy of risk exposure estimates (C58) |
| using their model (C59) | average increase in annualized alpha for funds with significant time-varying exposures (C22) |