A multivariate Markov model for simulating correlated defaults

Masaaki Kijima, Katsuya Komoribayashi, Eisuke Suzuki

Abstract

ABSTRACT

Recently, it has become common to use a Markov model to describe the dynamics of a firm’s credit rating as an indicator of the likelihood of default. However, when evaluating the credit risk of a portfolio consisting of credit-sensitive assets such as corporate bonds, existing Markov models cannot be applied directly because they do not account for default correlations. This paper proposes a multivariate Markov model for simulating the dynamics of correlated credit ratings of multiple firms. Our model is based on the single index model in modern portfolio theory and is an extension of Jarrow, Lando, and Turnbull (1997) to the multivariate asset case. The parameters of the model are estimated from observable data. The model will prove useful for credit risk management and for the pricing of credit derivatives of the basket type.