摘要
应用GARCH模型对深证成指日收益率从1991年4月3日到2006年3月12日共3 695个数据进行了拟合,结果表明GARCH模型能够很好地拟合日收益率时间序列,拟合后残差序列的均值、标准差、偏度、峰度、J-B统计量等都有明显改善,同时也发现波动具有明显的持续效应.
The GARCH models are adopted to forecast the variance of the daily benefit of ingredient indexes of Shenzhen stock market from April 3rd. 1991 to March 12th. 2006. The results obtained show that the GARCH model is a better model for time series of the daily benefit. The standard deviation, skewness, kurtosis and Jarque-Bera etc. of the residuals are improved a lot. The results also show that the persistence in volatility is evident.
出处
《北京师范大学学报（自然科学版）》
CAS
CSCD
北大核心
2006年第6期646-648,共3页
Journal of Beijing Normal University(Natural Science)
关键词
深证成指
波动性
GARCH模型
持续性
ingredient indexes of Shenzhen stock market
volatility
GARCH model
persistence
