Hedge Fund Risk Modeling

Apr 1, 2007

This paper introduces the Barra Hedge Fund Model, designed to overcome the unique challenges of modeling hedge fund risk.  The model provides a forecast of the risk of a hedge fund, or a portfolio of funds, using fund return series and information regarding its peer group membership.  Extensive research has identified factors that drive the returns to securities within each of these peer groups for various asset classes and regions.  What may be surprising to some investors is that many hedge fund managers do not fully hedge their exposure to these factors.  The model uses factors that identify two major sources of hedge fund systematic risk.  A portion of hedge fund risk is due to exposures to familiar factors that underlie conventional investments.   The hedge fund strategy factors capture systematic risk characteristics not fully explained by these traditional factors.   Each fund's exposure to these risks is calculated using a returns-based analysis.  We introduce the Barra hedge Fund Exposure Generator, an adaptive framework for estimating dynamic strategy exposures based upon moving and expanding window regressions, as well as a Kalman filter.  In addition, we address how to select the best factors for modeling a single hedge fund using in-sample and out-of-sample statistical analyses.  Most importantly, this approach can be used to aggregate factor exposures, factor covariance and idiosyncratic risk into a single portfolio risk forecast.  Lastly, we illustrate typical hedge fund style peer group systematic behavior and explain the drawbacks of using hedge fund style indices to model risk.