基于Copula函数的股票市场风险溢出网络特征研究  

Research on Risk Spillover Networks among Major Global Stock Markets Based on Copula Function

作　　者：任英华[1] 赵婉茹 罗良清[2] REN Ying-hu;ZHAO Wan-ru;LUO Liang-qing(College of Finance and Statistics,Hunan University,Changsha 410079,China;College of Statistics,Jiangxi University of Finance and Economics,Nanchang 330013,China)

机构地区：[1]湖南大学金融与统计学院,湖南长沙410079 [2]江西财经大学统计学院,江西南昌330013

出　　处：《统计与信息论坛》2020年第8期53-63,共11页Journal of Statistics and Information

基　　金：国家社会科学基金一般项目“复杂网络视角下系统性金融风险统计监测研究”(19BTJ024);湖南省自然科学基金面上项目“基于复杂网络的金融市场风险传染机理与时空效应研究”(2019JJ40038)。

摘　　要：为捕捉全球主要股票市场间时变风险溢出的网络结构与特征,识别各国在全球风险溢出中的角色与地位,把握风险溢出的全貌,采用时变Dcc-Garch-Copula-ΔCoVaR模型测度任意两两股票市场间的时变风险溢出,并基于小波多分辨分析提取风险溢出的特征量,构建静态与动态复杂网络进行分析。研究结果发现,时变Dcc-Garch-Copula-ΔCoVaR模型能够较好地捕捉股票市场的风险溢出演化趋势。在全局静态网络中,美国以及部分欧洲地区是风险溢出的中心,欧洲国家的金融风险传染具有明显的非对称性。美国、英国以及中国香港等国家或地区具有“主溢出”效应;中国、新加坡、泰国等发挥“经纪人”作用。动态风险溢出网络的聚类系数与平均路径长度基本呈反向变动,符合小世界特征。In order to capture the network topology properties of time-varying risk spillovers among the global stock markets,identify the role and status of countries in global risk spillovers,and grasp the whole picture of risk spillovers.Dcc-Garch-Copula-ΔCoVaR model is used to measure the time-varying risk spillovers among the global stock markets.Then the risk spillover networks is constructed from a static and dynamic perspective based on wavelet multi-resolution analysis.The results show that time-varying Dcc-Garch-Copula-ΔCoVaR model can better capture the evolution patterns of risk spillovers among the stock markets.In the global static network,the United States and some European countries are the centers of risk spillovers,and the financial risk contagion of European countries has obvious asymmetry.The United States,the United Kingdom and Hong Kong have a"net spillover"effect.China,Singapore,Thailand and others play the"broker"role.In dynamic networks,the clustering coefficient and average path length have opposite trends,which is consistent with the the"small world"character.

关 键 词：股票市场 时变风险溢出网络 Dcc-Garch-Copula-ΔCoVaR 小波多分辨分析 复杂网络 

分 类 号：F222[经济管理—国民经济] C812[社会学—统计学]