Application Of Copula-based Quantile Regression And VaR Method In Mainland China And Hong Kong Stock Markets

Investors tend to increasingly focus on effective risk measurement and control with the further opening of Chinese financial market and the development of Shanghai-Hong Kong Stock Connect and Shenzhen-Hong Kong Stock Connect.VaR is the most widely used measure of portfolio risk,while Copula theory can break down the portfolio risk into individual asset risk and correlated structure between assets.The thesis takes Shanghai,Shenzhen and Hong Kong stock markets as research objects,and selects daily closing price data in Shanghai Composite Index,Shenzhen Stock Index and Hang Seng Index from January 4th 2010 to December 12 th 2016.Research is conducted using Copula-based quantile regression and VaR method.First,the thesis will investigate the pairwise correlation between the Shanghai Composite Index,the Shenzhen Component Index and the Hang Seng Index from the perspective of Copula-based quantile regression.Special attention will be paid to the tail correlation of the market return rate variables.By fitting the Copula function and conducting tests,it is found out that t Copula,survival BB7 Copula,and BB7 Copula are better options for Shanghai Composite Index / Shenzhen Component Index,Shanghai Composite Index / Hang Seng Index,and the Shenzhen Component / Hang Seng Index respectively.A further step taken to perform quantile regression by adopting the above-mentioned three Copulas contributes to the finding that the interdependence among assets is present in any given quantile,including some extreme cases.Copula quantile regression could provide a more comprehensive way to measure the correlation between assets.Second,the thesis calculates the portfolio VaR by means of Copula-based function.By constructing pairwise equal weights portfolio among the three stock markets,the VaR obtained under this methods has less failure days and more significant effects than those by historical data simulation.