The impact of different correlation approaches on valuing credit default swaps with counterparty risk

This paper has two main contributions. We first build a general framework for valuing credit default swaps (CDS) with counterparty risk. We extend the work of Hull and White, and Hamp, Kettunen and Meissner, and build two quadruple trees. One tree represents the CDS spread payments, and one tree represents the CDS payoffs in the case of default. The trees are then combined using swap evaluation techniques to derive a closed-form solution for the CDS spread including counterparty risk. This is our first contribution. Our second contribution is testing the impact of five different reference asset--counterparty dependence concepts on the CDS price. We find that different dependence approaches lead to significanly different CDS spreads. The basic version of the model can be calibrated to market CDS spreads. A Visual Basic source code of the model can be provided upon request.