Research On Credit Risk Models Based On The Negative Correlation Assumption Between Default Probability And Recovery Rate

Credit risk, also called default risk, is the distribution of financial losses due to unexpected change in the credit quality of a counterparty in a financial agreement. Its range includes agency downgrades, failure to service and liquidation. Along with the rapid progress of corporate bond market, the global relaxation of financial restriction, and the continuous coming forth of new financial derivation, people know the importance of credit risk in the asset pricing and risk management more and more. As a result, a large number of models about credit risk have been built. The most outstanding representation is the continuous improvement of credit risk management in old and new Basel committee. Because china has jointed the WTO, the financial institutes whose competition-ability are weak, and have heavy cloth-wrappers, such as banks, will face more severe competition. At the same time, although the corporate bond market of our country grows rapidly, it's also lagging if comparing with the developed country. So, the research about credit risk has important theoretical and practical meaning to improve the competition of out country's finance institute, and to develop the bond market.Based on the comprehensive reading about the correlative literature, this dissertation makes a deep research about the existing credit risk models in chapter 1 and 2. Because the reduced-form models can represent the unexpected default more precisely and be consistent with empirical researches more accurately than tructural models, the credit risk pricing models of this dissertation belong to reduced-form models. According to the problem that the recovery rate is traditional treated as a constant or an independent stochastic variable by the classical credit risk pricing and management model, and problem that the negative correlation between the default probability and recovery rate is always neglected, this dissertation gets the exponential and logarithm regression models of default probablilty and recovery rate based on some empirical researches, and improves on several broadly applied credit risk models, such as structural hazard rate model, affine structure model, convertible bond pricing model and Credit Metrics model, and introduce the negative correlation between...