Extended Model Based On Vine Copula Applied In Financial Market

With the acceleration of the process of global economic integration,the systemic risks and uncertainties of the domestic financial market are also increasing.the risk of a market or financial institution may spread to other markets or financial institutions and even to the whole financial system through various channels,and the relationship between financial markets is becoming more and more complex.Vine Copula method has the ability to describe nonlinear and asymmetric dependency structure that traditional correlation measurement methods do not have.Therefore,the related research based on Vine Copula in the financial field is developing rapidly.This paper mainly uses the Vine Copula method to study the correlation measurement in the financial field,that is,the correlation estimation of the volatility of financial assets and the optimization of investment risk portfolio.Aiming at the research on the volatility of financial assets,this paper mainly discusses the estimation method of HAR(Heterogeneous Auto Regressive)expansion model based on Vine Copula and carries on the prediction analysis based on the conditional expectation of historical data.Based on the traditional volatility model(HAR family model)of high frequency data,this paper uses Vine Copula to jointly model the volatility components in different scales,and constructs an unstructured volatility model.In the empirical analysis,the high-frequency trading data of 20 stocks are collected as research samples,and different modeling schemes are used to compare the performance of the model.The results show that with the help of the advantages of Copula,the restriction on the model structure is relaxed,the limitation of the linear form of HAR family model is overcome,and the more complex dependence relationship between different volatility components is captured,so as to achieve a more accurate prediction effect.Aiming at the research of investment risk portfolio,this paper mainly discusses the estimation and prediction of COPAR(Copula Auto Regression)model for macroeconomic factors,and applies it to the investment risk portfolio model.Based on the Mean-CVa R model,this paper considers the influence of macroeconomic factors on the stock market,uses the dynamic factor model to reduce the dimension of macroeconomic factors,and uses the COPAR series model based on Vine Copula to capture the dependence of macroeconomic factors,and constructs the investment risk portfolio model based on COPAR.The empirical analysis collects the return data of 34 macroeconomic indicators and 6 stocks,applies the Dynamic Factor model and COPAR to the macroeconomic indicators,and applies the macroeconomic factors to the MeanCVa R model.The results show that considering macroeconomic factors can effectively reduce investment risk,and the use of COPAR model can better describe the complex dependence of the macroeconomic environment,accurately predict the volatility of future returns in the stock market,and achieve better portfolio performance.