Gamma process dynamic modelling of credit

The existing generation of credit derivatives models is unsatisfactory because they generally contain arbitrage, cannot describe the dynamics of the process, and are hard to extend beyond vanilla products. Martin Baxter has created a new tractable family of credit models, based on the gamma process, which allows arbitrage-free pricing of correlated credits in a dynamic model framework that can straightforwardly handle bespoke baskets and exotic products

The standard Gaussian copula model, with its overlay of base correlation, is useful but not ideal. It is essentially a static look-up table that does not model the dynamics of the process. It is hard to extend to bespoke baskets or other products, and it readily admits arbitrage.

Various models have been proposed to address these problems. These include both structural models that drive the value of the firm (or proxy) and reduced-form models that drive default intensity rates or the loss