摘要
运用时间序列分析的预测方法,对四大银行的股票日对数收益率序列进行拟合与预测分析,分别构建ARMA模型、GARCH模型以及ARMA-GARCH组合模型,通过模型比较,实证分析表明:在拟合效果上,ARMA-GARCH模型的拟合优度优于ARMA模型和GARCH模型;在预测效果上,ARMA模型的预测效果最优,ARMA-GARCH模型次之.
The prediction method of time series analysis was used to fit and predict the daily logarithmic return se-quences of China's four big Banks.ARMA model,GARCH model and ARMA-GARCH model were constructed respectively.Through the model comparison,the empirical analysis results show that,in terms of fitting effects the goodness of fit of AR-MA-GARCH model is better than that of ARMA model and GARCH model;and in terms of prediction effect,ARMA model has the best prediction effects followed by ARMA-GARCH model.
作者
李雄英
陈小玲
曾凯华
LI Xiongying;CIIEN Xiaoling;ZENG Kaihua(College of Statistics and Mathematics 9 Guangdong University of Finance and Economics Guangzhou, Guangdong 510320 ,China;Guangdong Research Center of Technological, Economic and Development,Guangzhou, Guangdong 510070,China)
出处
《经济数学》
2018年第4期21-27,共7页
Journal of Quantitative Economics
基金
广东省哲学社会科学"十三五"规划项目资助(GD16XYJ16)
广东省教育厅青年创新人才类项目资助(人文社科)(2016WQNCX046)
广东省高校优秀青年教师培养计划资助(YQ2015077)
广东省自然科学基金资助(2015A030313623)
