In this article, we computed the average alpha of 651 equity long/short hedge funds. We found that, on average, the average annualised alpha per equity long/short hedge fund is 8.8% and the average annualised alternative beta premium per equity long/short hedge fund is 5.0%. This means that on average, the alpha is higher than the alternative beta, during the period 2002 to 2007.
Some hedge funds market themselves as alpha providers and others as alternative beta providers. Alpha and beta are important for investors because they provide investors with what they are looking for – returns. By definition, alpha is the part of the return unrelated to any risk premium. The difficult part of the exercise is to separate alpha from beta.
We use a 24-month rolling window regression model:
where
a : monthly alpha
ß : exposure to the economic factor F
F : economic factor values
e : unexplained value
N : number of economic factors
The regression technique is an enhanced Sharpe-based style-analysis, with a 24-month rolling stepwise regression, where the beta must be significant with 90% probability. Using this regression technique, we measure the significant exposure between the hedge funds and seven economic factors: 1) MSCI EM Asia, 2) MSCI Small Caps EMU, 3) Russel 1000 Value, 4) Russel 1000 Growth, 5) Russel 2000, 6) CS High Yield Index, (CSHY Index) and 7) BXM Index (BXM Index, covered call on S&P500). The equity long/short hedge fund samples were taken from the Eurekahedge and HedgeFund.net databases. Only funds with US$ returns available between the period January 2002 to May 2007 were included (by using this technique, the study is exposed to survivorship bias). Hedge funds from the same company, but with different currencies or similar names have been eliminated. Some funds of funds wrongly classified as equity long/short hedge funds have also been removed. In order to account for hedge funds which have produced extreme returns, the hedge funds with the lowest and highest annualised alpha and lowest and highest beta premium (ie four hedge funds in total) have been excluded from the sample. After this screening, 651 equity long/short hedge funds were left.
Figure 1 below shows that the annualised alpha distribution is positively skewed. This means there are more alphas to the right of the mean (which is 7.3% per annum) than to the left. In other words, some hedge funds delivered very large positive alphas. Exhibit 1 presents the average alpha for each hedge fund.
The average annual alpha is 8.8%.
On average, the 24-month rolling regression model explains 43% of the 651 equity long/short hedge funds (ie the adjusted R2 equals 0.43, on average for each hedge fund, for each rolling regression of 24 months).
Figure 1: Annualised Alpha Distribution for 651 Equity Long/Short Hedge Funds
Figure 2 below presents the annualised alternative beta premium for the 651 equity long/short hedge funds. The alternative beta premium is the part of the hedge fund returns that can be explained by the seven factors listed above. The beta premium is positively skewed, which means some hedge funds have delivered high annual returns by being exposed to long-only indices: 40% of the hedge funds have significant exposure to small cap indices, 32% to MSCI Asia Index, 15% to a value and growth index, 17% to CS High Yield Index and 10% only to Covered Call S&P500 Index.
The average annualised beta premium is 5.0%.
Figure 2: Annualised Beta Premium for 651 Equity Long/Short Hedge Funds
Conclusion
Among the 651 equity long/short hedge funds with at least 77 monthly returns available, on average, the annual alpha equals 8.8% and annual beta premium 5.0%. The beta premium is mainly generated by exposure to small caps, to Asia and to value-growth indices. On average, 42% of the equity long/short hedge fund returns can be significantly explained, with long-only indices.
| Average for 651 Equity Long/Short hedge funds | |
|---|---|
| Annualised beta premium | 5.0% |
| Annualised alpha | 8.8% |
| Annualised return | 13.8% |
Note:
The software platform used in this paper is from www.AlternativeSoft.com. This software platform is able to measure alpha and alternative betas, as presented in this article. With this and many other features, the AlternativeSoft’s software platform is well suited for hedge funds selection.
