A Comparative Analysis of CDO Pricing Models under the Factor Copula Framework

Abstract

We compare some popular CDO pricing models, related to the bottom-up approach. Dependence between default times is modeled through Gaussian, stochastic correlation, Student t, double t, Clayton and Marshall–Olkin copulas. We detail the model properties and compare the semi-analytic pricing approach with large portfolio approximation techniques. We study the independence and perfect dependence cases and the uniqueness of base correlation. The ability of the models to fit the correlation skew observed in CDO market quotes is also assessed. Finally, we relate CDO premiums and the distribution of conditional default probabilities which appears as a key input in the copula specification.