The Analysis Of The Volatility Of Stock Market And The Dependence Analysis Among Stock Market, GDP And Exchange Rate

After the development for 20 yeras, stock market of China has had a considerable scale, and has become one of the most important investment areas for a large number of investors. China's stock market shows strong volatility. And the rule of the volatility has great particularity. Ony if fully understanding the volatility of the stock market, we can make better use of it for the market economy. China's economy has rapidly grown for about thirty years. GDP increased at the average rate of more than 9% every year. In July of 2005, the exchange system of RMB was reform, no longer only pegged to the dollar but started to refer to a package. Frome then on the exchange rate of RMB against dollar keeped changing, and RMB began to appreciate.The interaction and impact among the development of stock market,economic growth and the appreciation of RMB are attracted more and more attention. Therefore, analysing the dependence among the stock market, GDP and exchange rate becomes particularly important.In this paper, we fit the volatility of stock market by building a new model of financial time series, and analyse the dependence among the stock market GDP and the exchange rate using Copula theory. Main tasks are as follows:First, we build the ARMA-TGARCH-M model to analyse the stock market of China. In this paper, in view of the autocorrelation and heteroskedasticity of the return series, we build an ARMA model at first, and then solve the question of heteroscedasticity matching the risk by the ARMA-TGARCH-M model, and give the method of the estimation of VaR basing on the ARMA-TGARCH-M model at last. We use actual data to build an ARMA-TGARCH-M model for the stock market of Shanghai and Shenzhen, and estimate the vale of VaR basing on the model.The result of Back-test indicates the estimate of VaR is comparatively ideal.Second, we explore to apply Copula functions to analyse the dependence of macro economic variables to provide some reference for economic decision-making.Using Copula functions to analysis dependence, we can capture the non-symmetri correlation and more important correlation at tail. Copula functions were fewly used in the analysis of macroeconomic variables.We use Copula functions to build the model among stock market, GDP and exchange rate in order to canculate dependence metrics, and to provide advice and for economic police in this paper.