A Research On Integrated Risk Measurement Of Market Risk And Credit Risk Of Chinese Listed Commercial Banks

With the rapid advance of financial liberalization, globalization and financial product innovation, banking risks become more diverse and complex, meantime these risks also influence each other, with a non-linear correlation. This diversity and non-linear correlation puts forward higher requirements on risk management, and the focus of the risk management gradually turns to risk integration. So, it has important academic and practical value to integrate risks with considering non-linear correlation among credit risk and market risk.First, this paper analyzes the theoretical basis for the mechanism of interaction between risks. Then, taking 13 Chinese listed banks as research object, this paper adopts Copula to describe the dependence structure between risks. Finally we use Monte Carlo simulation to estimate the VaR of the risk portfolio. Through the above process, we integrate credit and market risk. In addition, when fitting the marginal distribution of risks, this paper compares the GARCH model and kernel density estimation in order to find which has a higher fitness.The empirical results show:(1) kernel density estimation has a higher fitness when fitting the marginal distribution of risks, and t-Copula function is the best one to describe the relationship between risks; (2) LR statistics of Kupiec test show that the model proposed to integrate risks in this paper is effective. Moreover compared with the traditional integration measure, the traditional method has a higher VaR, proving that there exists risk diversification effect. This provides a theoretical basis for banks to improve capital utilization; (3) the result of VaR under different ratios of credit and market risk shows that the greater credit risk weights, VaR is lower, that is to say, increasing credit risk weight is more appropriate from the current ideal state.