The Study Of Integrated Risk Measurement Based On Copula

With the rapid development of economic and financial globalization, financial market risk become more and more complex,and it is changing from the single one to the diversification,so it is necessary to integrate risk measurement.Now,integrated risk management model has been highly valued by industry and academia.The core of integrated risk management model is the integration management of different types of risks,such as market risk,operational risk and credit risk.At present,the financial institutions face many risk including credit risk,market risk and operational risk, the correlation between risks has the characteristics of nonlinear and tail dependence.The traditional Pearson correlation coefficient analysis method is under the assumption of normal distribution and no longer applicable.However,the Copula function is not unrestricted by the choice of marginal distribution and it can be better to describe the related structure of risk.Therefore,this paper systematically studies the application of Copula function in the integration of financial risk measurement.In theory part,we summarizes the related theory knowledge about Copula firstly,then thorough analysis the correlation of common used Copula,and discusses how to estimate parameters and select models for Copula model.Finally,summarized risk measurement indicators and its measuring method.In empirical part,in the view of the fact that integrated risk measurement must be in full consideration of the correlation of risks,this paper measure the integrated risk from two aspects which are the correlation and integration measurement of risk.The first empirical analysis the correlation of credit spreads and market risk,it is the study of correlation structure using the method of estimation parameters of Copula based on correlation coefficient.The empirical results show that there was a positive correlation between the two,and the relevant structure can be better described by the Frank Copula function.In the second case,we take12Chinese listed commercial banks as the research object.Firstly,We determine the distribution of return rate of each risk,and then construct the dependence structure using Copula,and select the best mode.Finally calculate the VaR and CVaR of different risk combination by using the Monte Carlo simulation method and the importance sampling algorithm,and compare the two algorithms by return test.The results show that the value at risk calculated by Monte Carlo is more efficient when the loss is small.The above work provides a better theoretical and technical support for the research of integrated risk management model in china.