Research On Credit Default Swaps Pricing Under Heterogoeous Beliefs

Since July 2007, the outbreak of the U. S. "subprimes" crisis, a large number of anomalies difficult to be explained by classical finance were revealed. Among them, the credit spread puzzle, excess volatility, arbitrage violation are the most striking ones. Because there is a tremendous shock to their credit risk management and credit derivatives pricing theory and practice of system activities, both academic and practice have given to their attention. These anomalies have become the focus of theoretical research.Different from the traditional financial model, this paper tries to introduce behavioral finance into the credit risk model of neoclassical finance, basing on the natural process of the generation of investor's belief. From the sources of heterogeneous behavior of individual in market micro-structure and following three natural processes of the generation of investor's belief:the external information's arrival, the initial subjective expectation and the information update and learning process respectively, we try to construct a credit risk model under heterogeneous belief by using noise traders model, irrational expectation model, Bayes' framework and empirical study, to analyze the impact of the heterogeneous factors on the credit derivatives prices, to explain the credit market abnormal phenomenon. Besides, with empirical model and actual data, we prove the rationality of the theoretical models. In all, the main research content and innovations areas are as follows:To begin with, staring from exogenous sources of heterogeneous beliefs - the accuracy of information, we build a credit risk model under the heterogeneous information. With this model, we can give credit the default swaps pricing and analyze the exogenous risk of heterogeneous information on equilibrium price of credit default swaps, where has been found that under equilibrium condition the more heterogeneous information, the more credit risk premium. In addition, we discussed the changing law between heterogeneous information and credit default swaps in non-equilibrium situation.Furthermore, with the results of existing models of credit risk, we deduce a credit risk model under irrational subjective expectations from the endogenous sources of heterogeneous beliefs - the subjective irrational expectations. We have got a different result from the Merton model and found that subjective factor in density pricing function is the cause of excess volatility and prices deviate from fundamental value, which explains the arbitrage violation in credit market.Last but not least, basing on the Bayesian framework and merging exogenous sources with endogenous sources of heterogeneous beliefs, we construct a pricing model of credit default swaps with learning behavior. Through theoretical analysis and graphics presentations, it displays that the joint role of heterogeneous factors increases the probability of default and negative skew of risk probability, the credit spread above the results of classical Merton model, which helps explain the "credit spread puzzle".Lastly, by summarizing the findings of theoretical model above and the previous literatures to be designed empirical variables, we tested the main conclusions of theoretical models through empirical analysis, found that after controlling company level variables (such as size, financial leverage), market liquidity indicators (such as stock volatility and option volatility), and macroeconomic factors (such as non-farm enroll, Fama-Frech three factors), the explanatory power of heterogeneous beliefs on the credit spread of the CDS is still significant.This paper follows the research path of theoretical model to model analysis, empirical hypothesis, then testing results to confirm each other, with results that heterogeneous beliefs is a risk factor of credit risk and credit derivatives pricing without including in traditional financial models and can help explain "credit spreads puzzle" which cannot be explained by the classical models. All findings in this paper can benefit actual credit risk management, help financial supervision to prevent financial crisis and improve risk management system of our commercial bank.