Abstract

In this paper, the risk-adjusted performance of dynamic asset allocation strategies for hedge fund indices, based on minimum variance and maximum Sharpe ratio approaches, is examined and compared to the S&P500 index benchmark. Furthermore, the added benefits of using conditional volatility forecasting, namely an asymmetric generalized autoregressive conditionally heteroscedastic (asymmetric GARCH) process, are examined when constructing dynamic hedge fund index portfolios. The evaluation period is based on a monthly out-of-sample comparison from May 2002 to June 2006 for nine Credit Suisse First Boston/Tremont hedge fund indices. Weekly and daily rebalanced dynamic portfolios are examined on the out-of-sample data from December 2005 until the end of June 2006 for the three main sub-indices of Standard & Poor's Hedge Fund Index. A multivariate asymmetric GARCH model is also considered for portfolio construction using daily S&P Hedge Fund sub-indices data. Before transaction costs are included, results show that when hedge fund indices exhibit volatility clustering, accounting for forecasted next-period volatility generates portfolios with the best risk-return profile among all portfolios under consideration. After accounting for transaction costs, out-of-sample results indicate that all dynamic hedge fund indices portfolios largely outperform the S&P 500 index, both on a risk-adjusted and nominal basis.