Content area
Full Text
Hedge funds (HFs) have been increasing in popularity over the last 15 years and still attract a great deal of interest among the finance community. The HF industry has been growing at an impressive rate. Indeed, the number of HFs has increased from 6,000 to 8,000 during the last ten years.1 The assets managed by HFs have skyrocketed over the last 16 years to reach $1.2 trillion at the end of 2005 with a growth of almost 3,000 percent.2 Research conducted by Van Hedge Fund Advisors International predicts that HF assets will at least double by 2009 ($2 trillion), quadruple to $4 trillion by 2013, and sextuple to $6 trillion by 2015.3
The spectacular growth of the HF industry has proved to be a fertile ground for research, which has come from both academics and practitioners. Researchers have mainly focused on the unique risk/reward profiles of HF investments, and the survivability and persistence of their performance over various periods (among others, Agarwal and Naik [2000]; Capocci and Hübner [2004]; Getmansky, Lo, and Mei [2004]; Jaeger and Wagner [2005]; and Malkiel and Saha [2005]); little is known, however, about the predictability of HF returns. This question is of interest given the ongoing debate about stock return predictability, in general, and, in particular, the evidence of Keim and Stambaugh [1986], Campbell [1987], Fama and French [1989], and more recently of Avramov [2004] and Guo [2006], that macro variables such as the dividend yield, the default premium, the term premium, and the consumption-wealth ratio forecast excess stockmarket returns. Additionally, Kandel and Stambaugh [1996] suggest that when returns are predictable, it is possible to have an economically significant impact on asset allocations, even though the uncertainty of future returns remains high. Kandel and Stambaugh's conclusion explains why investors continue to make tactical decisions about portfolio composition. Despite the abundance of studies examining stock return predictability, little evidence is available on the predictability of HF returns. Amenc, El Bied, and Martellini [2003] are the first to use multifactor models to forecast the returns of nine HF indexes and find strong evidence of significant predictability in HF returns. They also find that the potential benefits of tactical style-allocation portfolios are large. In this article, we extend their work by first...