基于Vine Copula的上证股指行业板块风险度量

The Risk Measurement of Shanghai Securities Industry Sectors Based on Vine Copula

摘要: 在2008年金融次贷危机的噩梦之后，系统性风险成为政策制定者和监管当局担忧的主要问题。文章在传统的Copula函数无法更好地解决“维数灾难”问题的情况下，引入藤Copula (Vine-Copula)，建立了能够更为灵活且精确地测度VaR的模型。在文中，首先，以上海证券交易所四个行业板块指数(工业指数、商业指数、地产指数、公共事业指数)为组合对象进行实证研究，对边缘分布用GARCH-st模型拟合。其次，基于不同的Vine-Copula结构构建Pair-Copula模型，通过极大似然值、AIC和BIC选择出合适的C藤结构Copula。最后，运用蒙特卡洛模拟方法计算多个投资组合的VaR，并通过Kupiec返回检验方法检验模型的有效性，以期为投资者和风险管理者提供更多借鉴。

Abstract: After the nightmare of the 2008 financial subprime crisis, systemic risk became a major concern for policymakers and regulators. In this paper, Vine Copula is introduced to build a more flexible and accurate VaR measurement model when the traditional Copula function cannot solve the problem of “dimension disaster” better. In this paper, first of all, taking the four industry sectors’ index in Shanghai stock exchange (industrial index, commercial index, property index, utilities index) as a composite object for empirical research, the marginal distribution is fitted by using GARCH-st model. Secondly, the Pair Copula model was constructed based on different Vine-Copula structures, and the appropriate C-Vine structure Copula was selected through maximum likelihood value, AIC and BIC value. Finally, VaR from a number of portfolios is calculated by using the monte carlo simulation method, and the effectiveness of the model is tested by Kupiec return test method, in order to provide more reference for investors and risk managers.