In this paper, we explain why a nonparametric approach based on a betakernel [Renault, Scaillet (2004)] will lead to significant bias when appliedto recovery rate distributions. This is due to a specific feature of thesedistributions, which admit strictly positive weights at 100 %...

We consider a quadratic stochastic intensity model with Gaussian autore-gressive factor, derive explicit formulas for the predictive mortality tables andprovide the recursive updating formulas are also provided. We also explainhow to use appropriately the Kalman ¯lter to estimate the parameters...

We develop a unified approach with closed-form solutions for pricing bonds, stocks,currencies and their derivatives. The specification assumes a fundamental risk factorrepresented by a stochastic positive definite matrix following a Wishart autoregressive(WAR) process. By assuming a...