Multiple Dynamic Risk Measurement Model Based On Vine Copula

VaR is a risk measurement method that is respected by major financial institutions and regulators.However,due to external shocks in financial markets and the impact of internal characteristics of data on financial markets and investors to reduce risk diversification,In the configuration,the VaR either underestimates the risk of the tail or neglects the state transition or does not accurately characterize the interdependent structure between the assets.Based on this,the main purpose of this paper is to establish a new portfolio VaR model which can accurately judge the fluctuation of sample data in each period and accurately describe the inter-asset-dependent structure to improve the forecast accuracy of risk.In this paper,two models are established in a step-by-step way.The first is a multi-dynamic VaR model based on SVt-EVT-Vine Copula.Firstly,the edge distribution model is established by combining the SVt model and the semi-parametric method of EVT model.(CVine,DVine and RVine)were used to construct the dependent structure model of financial time series,and the results were selected according to the AIC and BIC criteria.Combined with the MCMC simulation method to calculate the portfolio VaR,and compared with other models,highlighting the advantages of the model.Secondly,on the basis of the previous model,the influence of state transition on VaR is considered,and the Markov switching is introduced into SVt model to establish SVt-EVT-Vine Copula portfolio VaR model based on Markov switching The Comparing with the previous model,it is shown that the portfolio of SVt-EVT-Vine Copula based on Markov switching is effective and the risk forecasting ability is stronger,and CVine is described as the interdependent structure of assets.more accurate.Finally,the main contents of this paper are summarized,and the further research contents are analyzed.