Tail Approximation for Credit Risk Portfolios with Heavy-tailed Risk Factors

by Krassimir Kostadinov of the Munich University of Technology

January 2005

Abstract: We consider a portfolio credit risk model in the spirit of CreditMetrics. The multivariate normally distributed underlying risk factors in that model are replaced by more general multivariate elliptical factors with heavy-tailed marginals, introducing tail-dependence. We consider a full-scale version of the model, i.e. we incorporate not only the default risk, but also rating migrations, credit spread volatility and recovery risk.

We derive an upper bound of the portfolio loss distribution, which is particularly accurate at high loss levels. Given the complexity of our model, we obtain this results using a mixture of analytic techniques and Monte Carlo simulation. We conclude with an approximation of VaR and a new method to determine the contributions of the individual credits to the overall portfolio risk.

JEL Classification: G11, G21.

AMS Classification: 62H99, 65C20, 91B28.

Keywords: risk contributions, portfolio credit risk, stock. approximation, tail dependence.

Published in: Journal of Risk, Vol. 8, No. 2, (Winter 2006), pp. 81-107.

Books Referenced in this paper:  (what is this?)

Download paper (262K PDF) 24 pages