摘要
构建ARMA-EGARCH-t-Copula模型,研究上证B股指数和深证B股指数相关关系。选取上证B股和深证B股指数的日收盘指数序列,采用ARMA-EGARCH-t模型进行拟合边缘分布,并判断其具有杠杆效应。利用静态Copula和时变Copula模型构建两序列间的相关性,比较发现,利用动态SJC-Copula探究上海与深圳两市间的相关关系更合理。根据两股指间相关关系,采用蒙特卡洛模拟其VaR值来刻画两者间风险。实证结果表明,上海和深圳间的下尾相关性要强于上尾相关性,且在同一置信水平下,不同权重组合的VaR值不同,在相同权重组合下,置信区间提高, VaR值增大。
The ARMA-EGARCH- t -Copula is constructed to study the correlation of Shanghai-B Index and Shenzhen-B Index.Firstly,the model of ARMA-EGARCH- t is used to construct marginal distribution of Copula function for the daily closing sequence of the Shanghai-B Index and Shenzhen-B Index,and the leverage effect is judged.The correlation between the two sequences is constructed by using static Copula and time-varying Copula models and it is found that the dynamic SJC-Copula is more reasonable to study the correlation between Shanghai and Shenzhen.Finally,according to the correlation between the two indexes,the VaR value of monte carlo simulation is used to describe the risk between the two indexes.The empirical results show that the lower tail correlation between Shanghai and Shenzhen is stronger than the upper tail correlation;at the same confidencce level,the VaR values of different weight combinations are different,and at the same weight combinations,the confidence interval rises and the VaR values increases.
作者
薛凯丽
卢俊香
XUE Kaili;LU Junxiang(School of Science,Xi′an Polytechnic University,Xi′an 710048,China;School of Economics and Business Administration,Xi′an University of Technology,Xi′an 710048,China)
出处
《纺织高校基础科学学报》
CAS
2019年第1期109-115,共7页
Basic Sciences Journal of Textile Universities
基金
国家自然科学基金(11601410)
陕西省自然科学基金(2017JM1007)
中国博士后科学基金(179130)
