Research On Risk Measurement,optimization And Spillover Effect Of Chinese Financial Market

The globalization of the world economy and financial integration have created opportunities for countries to become increasingly open and integrated,and have achieved common development of the global economy.As an important role of the world economy,Chinese financial industries have also begun to shift from the main business to comprehensive management,and gradually formed development pattern of financial industry with diversified business.However,the development model of integrated management not only enhances the interdependence among financial industries,but also contributes the contagion of financial risks among industries,which makes the characterization of the relationship between financial assets more complex.Therefore,how to describe the complex interdependent structure among financial assets,accurately measure financial market risk and the risk spillover effects of various financial industries by scientific and effective methods have aroused widespread concern of financial researchers,regulators and investors.It has important theoretical and practical significance for risk management,investment decision-making and capital allocation,and is also an important foundation for the socialist economic and financial construction with Chinese characteristics in the new era.This paper takes the financial institutions of the banking,securities and insurance industries in the Chinese financial market as the research object.Firstly,we study the Granger causality and interdependence among the three industries by constructing the VAR-vine copula model.Secondly,based on fully considering the industry characteristics of different institutions in financial market,this paper proposes the vine copula grouped model to describe the complex interdependence among financial institutions from the banking,securities and insurance industries,and examines the impact of this model describing the dependent structure on risk measurement,optimization and spillover effects,respectively.At the same time,the accuracy of the model describing the interdependence among variables is inferred from the precision perspective of financial risk measurement,optimization and spillover effects.The main research work of this paper includes the following four aspects.Firstly,for the multivariate variables with Granger causality and dependence,this paper introduces the vine copula model into the VAR-copula model and constructs the VAR-vine copula model,which effectively overcomes the limitations of the binary copula function in characterizing the nonlinear dependence of multivariate variables with Granger causality.The Granger causality among banking,securities and insurance industries is analyzed,and the complex interdependence among the three industries is explored.Then the transmission direction and correlation degree of return volatility among different industries are comprehensively and systematically investigated.Secondly,this paper considers that binary copula grouped model is only applicable to characterizing the interdependence between two industries that each contains two financial institutions respectively.In fact,the banking,the securities and the insurance industry respectively contain multiple financial institutions.Therefore,this paper proposes a vine copula grouped model,which depicts the interdependence among financial institutions with different industry characteristics in more detail,offers the concrete steps of the new model algorithm and proves the validity of the algorithm.Finally,by the backtesting analysis method,it undergoes a comparative research on the precision of Chinese financial market risk measurement using the vine copula grouped model and the traditional vine copula model.The accuracy and validity of vine copula grouped model in financial market dependence modeling and risk prediction are confirmed from the perspective of empirical analysis.Thirdly,the accuracy of Va R is verified in the research of the existing financial risk optimization models.The coherent risk metric ES for complementarity of Va R is added and combined with the vine copula grouped model,then a mean-ES model based on vine copula grouped is constructed.The influence of the model describing the dependence structure among variables and the selection of risk metrics on the optimization of financial risk are considered,and unconditional coverage test and bootstrap method are used to test the minimum Va R and the minimum ES.Finally,empirical research shows that the minimum risk based on vine copula grouped mean-ES model is more accurate,and also verifies the superiority of vine copula grouped model in optimizing financial risk.Fourthly,in view of the fact that the existing research neglect the impact of dependent structure of assets in different industries on the study of risk spillover effect.This paper constructs the vine copula grouped-CoVaR model to study the risk spillover effects among banking,securities and insurance industries,and the risk spillover effects of various industries on the financial system,thus reveals the risk spillover effects of different financial industries.The backtesting results of CoVaR show that the vine copula grouped model can effectively improve the estimation accuracy of risk spillover effects of various industries on the financial system.Meanwhile,the accuracy of the model is verified from the perspective of risk spillover effects.