CAT Bond Spreads via HARA Utility and Nonparametric Tests

Abstract

After deriving a new utility-based model for pricing catastrophe (CAT) bonds under hyperbolic absolute risk aversion (HARA), the authors propose two specification tests that use nonparametric estimation techniques to test simultaneously for all possible misspecifications. Existing pricing models, including this new one, are then estimated and tested using CAT bond primary market data. The utility-based model they propose not only is well-suited for explaining the risk–return relation observed in the CAT bond market but also delivers the best performance among the tested models. The authors also provide new empirical evidence that the aggregate utility function of CAT bond investors exhibits decreasing absolute risk aversion.

TOPICS: Fixed income and structured finance, statistical methods, performance measurement