Searching for hedge fund alpha returns across the quantiles

Friday, October 9, 2015: 9:20 AM
Gregory Koutmos, Ph.D. , School of Business, Department of Finance, Fairfield University, Fairfield, CT
Johan Knif, Ph.D. , Hanken, Vaasa, Finland
Dimitrios Koutmos, Ph.D. , Worcester Polytechnic Institute, Worcester, MA
There is a growing body of literature dealing with the ability of hedge funds to earn alpha returns, known also as absolute returns. The models used in most cases are the market model and the Fama-French factor model and the methodology is usually the basic regression analysis. A significant limitation of standard regression analysis is that it focuses on the conditional mean of the distribution of returns and it ignores the tails of the distribution. It is possible for example that alpha returns may not be constant across the conditional distribution. 

This paper contributes to the discussion on conditional dynamic alphas by taking a different empirical approach. Instead of explicitly modeling the risk adjusted returns over time we allow the risk adjusted returns to depend on the conditional residual return distribution using a quantile regression approach. Implicitly, if the risk adjusted return is varying over the conditional return distribution it is expected that a realized performance measure like alpha will explicitly exhibit a nonlinear and time-varying behavior. The advantage of this approach is that there is no need to explicitly define either the economic state variables or specify specific models for the time varying behavior. Furthermore, the paper also compares the estimated performance alphas of the quantile regression with those estimated by traditional OLS and Garch-type techniques.

More specifically this study aims to provide answers to the following questions:

1) Do hedge funds earn alpha-returns and which styles are more successful in this regard?

2) Are alpha-returns asymmetric across the conditional distribution?         

3) Are alpha-returns and specific risk factor exposures stable across the conditional distribution?

Data:

That data used in this study are monthly returns on hedge funds indices with the following investment styles:

Hedge fund index, Convertible arbitrage, Dedicated short bias, Emerging markets, Equity market neutral, Event driven, Event driven distressed, Event driven multi-strategy, Event driven risk arbitrage, Fixed income arbitrage, Global macro, Long/Short equity, Managed futures, Multi-strategy.

The data cover the period  1994:1 till 1//2015 for total of 253 monthly observations. The return series data are obtained from the TASS Hedge Funds Data Base which produces indexes of investment performance of several hedge fund classes.