本文结合SV模型和Copula技术,建立两变量金融时间序列的Copula-SV模型,并以上海综合指数和深圳成分指数为例利用建立的模型进行分析,根据采用不同的Archimedean Copula函数,通过使用K-S检验说明用Clayton Copula研究上证综指和深圳成指的下尾相关性,用Gumbel Copula研究上证综指和深圳成指的上尾相关性是合适的,从而风险管理者就可以根据尾部相关性,定量的研究两个市场的相关性及预测市场的变化。
This paper sets up a bivariate financial time series model combined with SV model using Copula technology and makes empirical analysis on the Shanghai Composite Index and the Shenzhen Component Index. Through using the K-S test, the results show that using Gumbel Copula to study the upper tail dependence of two indexes is appropriate and using Clayton Copula to study the lower tail dependence of them also works. Based on the results, the risk managers can do some empirical research and forecast the market changes.