上证380高频指数数据已实现GARCH(1,2)模型的风险测量  

Measurement of Risk Based on Realized GARCH(1,2) Model with Different Residual

作　　者：魏正元[1] 张鑫[1] 赵瑜[1] 

机构地区：[1]重庆理工大学数学与统计学院,重庆400054

出　　处：《重庆理工大学学报（自然科学）》2015年第5期137-141,共5页Journal of Chongqing University of Technology：Natural Science

基　　金：重庆市自然科学基金资助项目(cstc2012jj A00018);重庆市教委科学技术研究项目(KJ130810);重庆市高等教育教学改革研究项目(1203053)

摘　　要：针对高频金融数据收益率序列的厚尾和偏斜性,建立了偏t误差分布假设下的R-GARCH(1,2)模型,对上证380指数5 min频率的高频数据进行了Va R预测,并与经典的正态分布和t分布误差假设下的R-GARCH(1,2)模型的预测精度进行了对比分析。结果表明,误差项服从偏t分布的R-GARCH(1,2)模型能够有效识别上证380指数收益率序列的分布特征,并且能够精确地测量其收益风险。We built the realized GARCH（ 1,2） model with skewed student＇ s t distribution to forecast VaR of Shanghai Stock Exchange 380 index for the high-frequency financial data with a fat tail and a- symmetry. The model with normal distribution and student＇ s t distribution was used for comparison. The empirical results show that the realized GARCH model with the skewed student＇ s t distribution can identify the characteristics of the return series of Shanghai Stock Exchange 380 index and measure the risk of Shanghai Stock Exchange 380 index more accurately.

关 键 词：高频金融数据 已实现GARCH VAR 偏T分布 厚尾特征 偏斜性 

分 类 号：O21[理学—概率论与数理统计]