Research On The Dependent Structure Of Unstable Stocks In Chinese Stock Market

China's economy is at a stage of rapid development,and this stage has greatly promoted the development of China's stock market.Due to its own characteristics,the unstable stocks have become the focus of investors in the stock market and the research focus of domestic and foreign scholars.Generally speaking,if you want to get more returns in a short period of time,you must pay more attention to the unstable stocks,because the unstable stocks can rise in a short period of time and get huge returns.However,it is known that the higher the return rate is,the higher the risk is.Compared with other stocks,the unstable stock has the characteristics of sharp peaks and thick tails,and it is more prone to extreme values and higher risks.Therefore,scientific and accurate financial decisions are especially important in the investment process.This article studies and analyzes the interdependent structure of the unstable stocks in depth,and measures the risk of the unstable stocks based on the grasp of the interdependent structure of the unstable stocks.Finally,it puts forward scientific and accurate suggestions for investors to invest in the unstable stocks.First of all,based on the research results at home and abroad,this paper systematically introduces the volatility model,Copula,Vine Copula's theory,properties,and its application in finance and summarizes the steps of constructing Vine Copula structure in detail.Introduce the value at risk(VaR),a comprehensive summary of the methods commonly used to calculate VaR,and specifically summarize the detailed steps of calculating VaR.Secondly,in the empirical analysis part,high-frequency data of 16 unstable stocks for five minutes were selected from9:35 on November 13,2018 to 14:50 on March 22,2019.An ARMA-EGARCH(1,1)marginal distribution model was established based on domestic and foreign literature and characteristics of unstable stocks' high-frequency data.Furthermore,use the Vine Copula model to explore the interdependence between unstable stocks,and theAkaike information criterion(AIC)value and Clarke test which specially used to compare the fitting effects of two different R-Vine models,were used to compare the fit of the R-Vine all family model,R-Vine all t Copula Model and R-Vine all Gaussian Copula Model.Finally,the risks of the unstable stocks are forecasted and the effects of predicting VaR based on the ARMA-EGARCH(1,1)model and the ARMA-EGARCH(1,1)-R-Vine Copula model are compared.Through the analysis,modeling and comparison of the unstable stocks,the ARMA-EGARCH(1,1)model can fit the high-frequency data of the unstable stocks for five minutes.The R-Vine all family model can better fit the dependency structure between unstable stocks than the other two models,which illustrates the complexity of the dependency structure of the tail of the unstable stocks.The single Copula function often cannot describe the complex structure of the tail of the unstable stocks.By comparison,we also get the ARMA-EGARCH(1,1)-R-Vine Copula model more accurately than the ARMA-EGARCH(1,1)model for predicting VaR;Finally,the analysis results of dependent Structure Model of unstable stocks and VaR are summarized and suggestions for investing in unstable stocks are proposed for investors.