Joint Survival Analysis of Hedge Funds and Funds of Funds Using Copulas

Abstract

This paper uses copulas to model the joint survival within the groups of hedge funds and funds of funds managed by the same manager. Given their skewed distribution, a simple survival analysis based on linearity assumptions may fail to fully capture the dependence caused by extreme events in the tails. The study employs both one and two-parameter families of copulas to capture more than one type of dependence, like upper and lower tail dependence and concordance measures. The paper uses both a one-step approach where the margins and the copula parameters are estimated jointly, and a two-step approach where the margins are fitted first and the copula parameter is estimated thereafter given the fixed margins. The results show that the survivor functions of hedge funds and funds of funds while different from each other are strongly dependent, with the degree of dependence increasing in the tails. The evidence in the paper aligns with previous findings on financial markets contagion and it shows that contagion risk is especially important between hedge funds and funds of funds. In particular, the results suggest that a simple linear approach may provide misleading results regarding the full degree of association between hedge funds and funds of funds.

Keywords: Hedge Funds, Funds of Funds, Copula functions, Survival analysis

JEL Classification: C130, G290