1 - Rank alpha funds of hedge funds

Abstract

This chapter examines the performance of portfolios of hedge funds when fund selection is based on the rank of a funds' alpha rather than on the estimated value of the alpha. It presents four different factor models to estimate the alpha of individual hedge funds. The base case model is the simplest representation of the fund returns as a function of the two most important underlying asset classes, equities and bonds; the broad fundamental factor model employs several indices to capture the performance of the main asset classes and other factors representing specific types of nonlinear strategies, such as market timing, volatility trading, and equilibrium trading; the multifactor model is based on hedge fund indices; and finally, the statistical factor model is based on factors extracted from fund returns through principal component analysis. For each of these four factor models, the hedge fund selection process is determined only by the rank of the fund's alpha and not by the actual value of its estimated alpha. As different factor models provide very different estimates of a hedge fund's alpha, so when portfolio optimization uses estimated alpha values there is a high degree of model risk. On the other hand, even for the simplest factor models one finds that ranking funds according to their alpha estimates is an efficient selection process. In an extensive out-of-sample historical analysis, funds of funds that are selected in this way and then allocated using constrained minimum-variance optimization are shown to outperform an equally weighted index of all funds, minimum-variance portfolios of randomly selected funds, and portfolios that are optimized using estimated alpha values. Of the four factor models presented, the best out-of-sample performance is obtained using the rank alphas from the principal components factor model.