基于半参数CARE模型的金融市场VaR度量  

The Estimation of Value at Risk in Financial Market Based on Semi-parametric CARE Models

作　　者：胡宗义[1] 李毅[1] 万闯 唐建阳 HU Zong-yi;LI Yi;WAN Chuang;TANG Jian-yang(School of Finance and Statistics,Hunan University,Changsha 410079,China;Gregory and Paula Chow Center for Economics Research,Xiamen University,Xiamen 361005,China))

机构地区：[1]湖南大学金融与统计学院,湖南长沙410079 [2]厦门大学邹至庄经济研究中心,福建厦门361005

出　　处：《统计与信息论坛》2019年第4期19-24,共6页Journal of Statistics and Information

基　　金：教育部规划基金项目<绿色发展中能源环境政策效应的动态CGE研究>(17YJA790030);湖南省研究生科研创新项目<绿色发展政策实施效果跟踪及评估研究>(CX2018B157)

摘　　要：以CAViaR模型为基础,结合Expectile模型,构建半参数CARE模型,度量金融市场的在值风险。选取2003年1月3日至2015年12月31日上证综合指数与深圳成份指数为研究对象,分别采用半参数CARE模型与GARCH模型刻画VaR的波动情况,并运用几类常返检验来评估模型的优劣。结果表明:GARCH模型能更好地刻画深证成份指数1%VaR与上证综合指数5%VaR,而半参数CARE模型能更好地刻画深证成份指数5%VaR与上证综合指数1%VaR。This paper propose semi-parametric CARE model,which is based on Expectile models and CAViaR models,to estimate VaR.based on the Shanghai composite index(SSEI) and Shenzhen component index(SZSEI) from January 3,2003 to December 31,2015,this paper uses semi-parametric CARE model and GARCH model depict the risk fluctuation respectively,and some backtesting analysis are taken to evaluate the models.Empirical analysis shows that,GARCH model is better in depicting 1%VaR in SZSEI and 5%VaR for SSEI.Meanwhile,semi-parametric CARE model is better in depicting 5%VaR in SZSEI and 1%VaR for SSEI.

关 键 词：Expectile 半参数CARE模型 VAR 风险度量 常返检验 

分 类 号：F224[经济管理—国民经济]