Pricing of corporate bonds and credit derivatives: Theory and evidence.

摘要

The risk of default occupies a central role in the pricing and hedging of credit risk. Different approaches to modeling default probabilities, and their migration, have been pursued. There are three main quantitative approaches to analyzing default risk as follows: the structural approach, the reduced-form approach, and the hybrid approach. This dissertation presents a new hybrid term structure model which can be used for pricing defaultable bonds and credit derivatives.; Specifically, I develop a four-factor defaultable term structure model for the pricing of corporate bonds and credit derivatives on the basis of the Duffie-Singleton Model. One of the factors that determine the credit spread is the firm-specific financial distress variable or index. The greater the value of the firm-specific financial distress index, the lower the quality of the firm. The second factor is the non-defaultable short rate which is assumed to follow the Cox-Ingersoll-Ross model or the Vasicek model, both of which are popular in the field of the term structure of interest rates. The third factor is the short rate spread, which is considerably driven by the firm-specific financial distress index and the default-free interest rate. The last factor is the volatility of the short rate credit spread. This model is tractable as well as flexible. Simulation results show that my model is capable of producing many different shapes for the term structure of credit spreads and captures many of the complex properties inherent in credit spreads.; In addition, I extend the work of Tahani (2004) and Longstaff and Schwartz (1995) by introducing both interest rate risk and stochastic volatility risk in pricing credit derivatives such as spread options. For the general mean-reverting process for the credit spread and two stochastic volatility processes: the square-root process and the Ornstein-Uhlembeck process, I derive the closed-form solutions for valuing European credit spread options using the Fourier inversion framework.; Finally, I develop Kalman filtering methodology to estimate the four-factor model and apply this to both Treasury and corporate bond indices.; The theoretical features of my model are well supported by the empirical analysis carried out. It is an interesting topic for further research to calibrate and test this new four-factor model using individual corporate bond data.