Computing Credit Valuation Adjustment for Bermudan Options With Wrong Way Risk

Abstract

We study the impact of wrong way risk (WWR) on credit valuation adjustment (CVA) for Bermudan options. WWR is modeled by a dependency between the underlying asset and the intensity of the counterparty's default. Two WWR models are proposed, based on a deterministic function and a CIR-jump (CIRJ) model, respectively. We present a nonnested Monte Carlo approach for computing CVA-VaR and CVA-expected shortfall (ES) for Bermudan options. By varying correlation coefficients, we study the impact of credit quality and WWR on the optimal exercise boundaries and CVA values of Bermudan products. Stress testing is performed.

Relevance of Wrong-Way Risk in Funding Valuation Adjustments

“…As seen in Section 3.4 for the IR swap, for some cases it is possible to find an analytical WWR approximation 6 . If this is desired, a portfolio of derivatives V allows for an analytic WWR approximation if it can be written in terms of y r (t, u), such that E t y l r (t, u) (V (u))…”

“…Literature on WWR modelling for CVA for non-credit derivatives can be divided into three approaches. 1 Firstly, there is simulation, which models interest rates, default intensities and their dependence through either a deterministic relationship or a set of correlated SDEs [6,10]. Secondly, a copula approach where the multivariate distribution of exposure and defaults is modelled through a copula [4,5].…”