基于Skew-t-GJRGARCH(1,1)模型的5G板块风险度量研究  

Research on Risk Measurement of 5G Plate Based on Skew-t-GJRGARCH(1,1)Model

作　　者：李世君 唐国强[1] 杜诗雪 LI Shijun;TANG Guoqiang;DU Shixue(School of Science, Guilin University of Technology, Guilin Guangxi 541006, China)

机构地区：[1]桂林理工大学理学院,广西桂林541006

出　　处：《广西师范大学学报（自然科学版）》2020年第5期56-63,共8页Journal of Guangxi Normal University:Natural Science Edition

基　　金：国家自然科学基金(61703117,61763008);广西中青年教师基础能力提升项目(2018KY0261)。

摘　　要：本文对5G板块进行风险度量研究,结果表明:对数收益率序列存在有偏、尖峰厚尾、非正态的特征,同时具有ARCH效应;通过拟合不同分布下的GJRGARCH(1,1)模型,发现5G板块不存在杠杆效应;采用GJRGARCH(1,1)模型在不同分布下进行VaR计算,由Kupiec检验得到基于skew-t分布的GJRGARCH(1,1)模型可以更好地度量5G板块的金融风险,其可信度和稳健性强。In this paper,the research on the risk measurement of the 5G plate is carried out.The results show that the logarithm of the yield series has a bias,a sharp peak,a non-normal feature,and an ARCH effect.The GJRGARCH(1,1)model with different distributions is fitted.It is found that there is no leverage effect in the 5G plate;the VaR calculation is performed under different distributions using the GJRGARCH(1,1)model,and the GJRGARCH(1,1)model based on the skew-t distribution can be better measured by the Kupiec test.

关 键 词：GJRGARCH 5G板块 skew-t VAR 风险度量 

分 类 号：O213[理学—概率论与数理统计] F224[理学—数学]