Abstract:
This paper presents a useful implementation of the Black and Cox (1976) structural credit risk models, and provides a distinct approach for estimation of model parameters. Distinguished from the rest of the literature, our approach offers two significant breakthroughs: one is estimating unknown model parameters based on observed CDS spreads, the other is introducing an extended Kalman filter into the credit risk sector. The extended Kalman filtering is an alternative to the particle filtering algorithm which is proposed by Duan and Fulop (2007). In our paper, some experiments are carried out, evaluating the performance of our approach. Finally, we demonstrate the application of our approach by using it to estimate a value process for General Motors from its CDS spreads.
