Hedge funds have alpha and beta. There are many models to evaluate the performance of hedge funds. In this article, we analyzed the performance of the overall hedge fund index as well as different styles of hedge funds based on the CAPM model. The time period we looked at for the whole dataset is from January 1994 to December 2010.
The Hedge Fund dataset used in this project is from Dow Jones Credit Suisse Hedge Fund Indexes, from 1994 to 2010. The index uses the Credit Suisse Hedge Fund Database, which tracks approximately 8000 funds and consists only the funds with a minimum of US$50M under management, a 12-month track record, and audited financial statements. The index is asset-weighted, is calculated and rebalanced on a monthly basis, and reflects performance net of all fees and expense. In addition to the overall index, the dataset also includes indexes by hedge-fund style. The styles used in this project are Convertible Arbitrage, Emerging Markets, Event Driven, Fixed Income Arbitrage, Global Macro, Long/Short Equity and Managed Futures. The market dataset used in this project is monthly market returns, 1-month Treasury bill rates, and monthly returns of the four empirical factors (Market, SMB, HML and MOM), from 1994 to 2010.
Below is the result of the regression under CAPM.
The table summarizes the regression results with t-stats in squared brackets. For the overall hedge fund industry, we see a non-trivial exposure to the market risk. The hedge fund index has a market beta of 0.29 and is statistically significant with a t-stat of 11.37. Compared with equity mutual funds, whose market betas are close to 1, this level of market exposure is relatively low and is characteristic of hedge funds. The R-square of the regression is 39%, indicating that the market portfolio can explain 39% of the random movements in the hedge fund index, which means that an important component of the risk in hedge fund returns comes from the market risk, although there is still a large component remained that can’t be explained by the market risk. Comparing this with the equity mutual fund industry whose R-squares are usually around 90%, we found that in hedge fund industry this unexplained component is relatively high. The overall hedge fund index has an alpha of 35 basis points per month under CAPM with a significant t-stat of 2.89, which means hedge funds on average provide positive performance under CAPM.
In addition to the overall hedge fund index, we also have the regression results for various hedge fund styles. The dedicated short hedge fund has a negative exposure to the market, as it has significant market beta of -0.84. Moreover, it has the highest R-square of 65.69% across all styles of hedge funds. In other words, although they tend to move against the overall market, the market can in fact explain a large component of their movements. The equity neutral hedge fund is not strictly market neutral, as it has a market beta of 0.18 and a significant t-stat of 4.24. This level of market exposure is relatively low when compared with Long/short equity hedge fund, whose market beta is 0.45 with a significant t-stat of 15.77. In addition, the long/short equity fund has the second highest R-square of 55.2% among all the hedge fund styles.
