Time-changed CIR default intensities with two-sided mean-reverting jumps

Abstract

The present paper introduces a jump-diffusion extension of the classical diffusion default intensity model by means of subordination in the sense of Bochner. We start from the bi-variate process $(X,D)$ of a diffusion state variable $X$ driving default intensity and a default indicator process $D$ and time change it with a Lévy subordinator $\mathcal{T}$. We characterize the time-changed process $(X_t^{\varphi },D_t^{\varphi })=(X({\mathcal{T}}_t),D({\mathcal{T}}_t))$ as a Markovian--Itô semimartingale and show from the Doob--Meyer decomposition of $D^{\varphi }$ that the default time in the time-changed model has a jump-diffusion or a pure jump intensity. When $X$ is a CIR diffusion with mean-reverting drift, the default intensity of the subordinate model (SubCIR) is a jump-diffusion or a pure jump process with mean-reverting jumps in both directions that stays nonnegative. The SubCIR default intensity model is analytically tractable by means of explicitly computed eigenfunction expansions of relevant semigroups, yielding closed-form pricing of credit-sensitive securities.

Published in at http://dx.doi.org/10.1214/13-AAP936 the Annals of Applied Probability (http://www.imstat.org/aap/) by the Institute of Mathematical Statistics (http://www.imstat.org)