上证指数周内效应研究——基于AR-GARCH-GED模型和滑动窗口回归
发布时间：2018-07-26 17:17
【摘要】：通过引入自回归项,改进了以往国内研究所使用的GARCH-GED模型,建立了AR-GARCH-GED模型,使用1990年12月19日~2015年7月15日期间将近25年的上证指数日数据研究其周内效应,得出了上证指数较之前研究更为显著的周内效应。为了进一步考察显著的周内效应是否仅仅是一个数据挖掘的纯机会主义行为,对该模型进行滑动窗口回归发现,周内效应显著的比例仅在13%~25%之间。基于该研究结果,国内以往研究中普遍存在的周内效应极有可能是一个统计假象,上证指数的周内效应依赖于数据挖掘。
[Abstract]:By introducing the autoregressive term, this paper improves the GARCH-GED model used in previous domestic research institutes, and establishes the AR-GARCH-GED model. Using the daily data of Shanghai stock index from December 19, 1990 to July 15, 2015, the intraweek effect is studied. The results show that the index of Shanghai stock market is more significant in the week effect than the previous study. In order to further investigate whether the significant intraweek effect is merely a purely opportunistic behavior of data mining, the sliding window regression of the model shows that the proportion of the significant effect in the week is only between 13% and 25%. Based on the results of this study, it is possible that the intraweek effect in previous studies in China is a statistical illusion, and the intraweek effect of Shanghai stock index depends on data mining.
【作者单位】： 重庆大学经济与工商管理学院;西南财经大学会计学院;
【基金】：国家社会科学基金重大项目“开放经济条件下我国虚拟经济运行安全法律保障研究”(项目编号:14ZDB148)
【分类号】：F224;F832.51
[Abstract]:By introducing the autoregressive term, this paper improves the GARCH-GED model used in previous domestic research institutes, and establishes the AR-GARCH-GED model. Using the daily data of Shanghai stock index from December 19, 1990 to July 15, 2015, the intraweek effect is studied. The results show that the index of Shanghai stock market is more significant in the week effect than the previous study. In order to further investigate whether the significant intraweek effect is merely a purely opportunistic behavior of data mining, the sliding window regression of the model shows that the proportion of the significant effect in the week is only between 13% and 25%. Based on the results of this study, it is possible that the intraweek effect in previous studies in China is a statistical illusion, and the intraweek effect of Shanghai stock index depends on data mining.
【作者单位】： 重庆大学经济与工商管理学院;西南财经大学会计学院;
【基金】：国家社会科学基金重大项目“开放经济条件下我国虚拟经济运行安全法律保障研究”(项目编号:14ZDB148)
【分类号】：F224;F832.51
【相似文献】
相关期刊论文 前10条
1 苟格;;一种基于上证指数的新方法研究[J];四川文理学院学报;2007年05期
2 於玮;;上证指数结构性变化的计量分析[J];金融经济;2007年22期
3 刘迎春;;沪深300指数与上证指数关系的实证研究[J];辽宁经济;2008年12期
4 姜娉婷;刘颖博;;上证指数与基金资产配置相关性实证研究[J];金融经济;2009年14期
5 朱宁;徐标;仝殿波;;上证指数的时间序列预测模型[J];桂林电子工业学院学报;2006年02期
6 范漪涵;;生存模型对上证指数涨跌天数的探讨[J];科协论坛(下半月);2007年03期
7 周晋;;从上证指数和沪深300的相关性看权重股的影响[J];上海工程技术大学学报;2008年04期
8 王强;秦安增;;上证指数短期预测的数学模型[J];电脑编程技巧与维护;2009年04期
9 栾s，
本文编号：2146719
本文链接：https://www.wllwen.com/jingjilunwen/touziyanjiulunwen/2146719.html
