基于时变Copula的La-VaR测度研究  

Measurement of Liquidity-adjusted VaR Based on Time-varying Copula

作　　者：江红莉[1] 何建敏[2] 胡小平[2] 

机构地区：[1]江苏大学财经学院,江苏镇江212013 [2]东南大学经济管理学院,江苏南京211102

出　　处：《重庆大学学报（社会科学版）》2013年第3期27-32,共6页Journal of Chongqing University(Social Science Edition)

基　　金：国家自然科学基金项目"基于复杂网络的银行间传染风险及其演化模型研究"(71071034);江苏大学高级技术人才科研启动基金项目(12JDG130)

摘　　要：目前关于流动性调整的市场风险测度研究,主要是静态模型。针对此,文章提出经流动性风险调整的市场风险动态测度的时变Copula方法。该方法使用连接函数构建流动性风险和市场风险的联合分布,能够兼顾这两种风险的非正态特征和它们之间的动态相关结构。基于该方法度量了中国股市经流动性调整的市场风险La-VaR,Kupiec检验表明,基于时变Copula模型预测La-VaR的效果优于基于常相关Copula模型的预测效果,并且时变T-Copula模型优于时变N-Copula模型。The research on liquidity-adjusted market risk is mainly based on static model at now.In this paper,the method of dynamic measurement of liquidity-adjusted market risk is proposed which is based on time-varying Copula.Time-varying Copula function is used to construct the joint distribution of liquidity risk and market risk,which can give attention to both non-normality of the two risks and their dynamic dependency.The liquidityadjusted market risk La-VaR of Chinese stock market is calculated.The Kupiec test shows that,the time-varying Copula is better than constant correlation Copula in the aspect of forecasting the value of La-VaR,and the timevarying T-Copula is better than the time-varying Normal-Copula.

关 键 词：市场风险 流动性风险 经流动性调整的市场风险 时变COPULA 极值理论 La-VaR 

分 类 号：F840.32[经济管理—保险]