A Study Of The Volatility Spillover Effect Based On Variable Structure Copula Model

Stock market is a complex dynamic system, economic globalization and financial liberalization are exacerbating the complexity and volatility of the stock market. Therefore, correlation of volatility between markets has been significantly enhanced. Volatility spillover effect between different financial markets refers to the volatility of the possible interaction; volatility will transfer from one market to another, making a market’s volatility to be restricted by other markets’ fluctuations. Therefore, in order to improve the accuracy and reduce the risk of decision-making in financial markets, a study of volatility spillover effect between financial markets is a crucial topic. Copula function can describe the dependence structure between multiple random variables; it is an effective tool to study the volatility spillover effect between different markets. In the empirical research, this paper uses the Bayes test to diagnose structure change points, uses copula function to obtain correlation coefficients of different markets in different stages, checks whether there is a significant difference, and analyze the existence of volatility spillover effect between different markets. This paper contains following aspects:(1) Test returns of stock markets residual time series’ARCH effect. Results show that the chosen stock markets’ time series have a significant ARCH effect under 5% significance level. Therefore, GARCH model can be applied.(2) Adopt GARCH (1,1)-t model to describe marginal distributions of chosen time series of stock markets’ returns. Initially, estimate the marginal distribution of model’s parameters, then use K-S test to inspect whether the time series which are from probability integral transform process obey (0,1) uniform distribution. The results indicate that return series are significantly leptokurtic, and Shanghai composite indexes are more likely to have extreme values when compared with other markets.(3) Apply static and dynamic copulas functions to analyze correlations between different stock markets. The static results show that the intensity of correlation of central Asian stock markets is the most strong. The dynamic results show that the degree of correlation between returns’series of Shanghai’s and other stock markets has a upward-sloping trend, especially since 2007, this trend is becomes more obvious.(4) Utilize three-stage method to study the volatility spillover effect between stock markets. The results indicate that the volatility spillover effects between returns’ series of Shenzhen and Shanghai stock markets are more obvious and occurred more frequently; the Shanghai and other four markets also exist volatility spillover effects at different times.(5) Using the variable structure model of copulas to research the volatility spillover effects between different stock markets conforms to the reality of the stock market, which proved the rationality of this method.