A simple nonparametric approach to pricing credit default swaps

Abstract

This study introduces a nonparametric approach to pricing credit default swaps (CDSs) and other single-name credit-risky securities. This method is notable for its simplicity, estimation speed, and flexibility. That is, it relies exclusively on closed-form solutions, which provide instantaneous results, and allows the user to reproduce any term structure of CDS spreads. I empirically assess its pricing performance by comparing it with an otherwise equivalent semiparametric (piecewise constant default probability) model that requires a series of root-search algorithms and represents the current market convention for marking-to-market CDS contracts. This analysis demonstrates that the new method also implies a reduction in mean percentage absolute pricing errors.