统计研究 ›› 2007, Vol. 24 ›› Issue (8): 88-91.

• 论文 • 上一篇    下一篇

局部平稳过程、稳定过程及GARCH模型的比较研究

李锐 向书坚   

  1. 中南财经政法大学信息学院统计系
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-08-15 发布日期:2007-08-15

Comparative Study on Local Stationary Process, Stable Process and GARCH Model

Li Rui ;Xiang Shujian   

  • Received:1900-01-01 Revised:1900-01-01 Online:2007-08-15 Published:2007-08-15

摘要: 本文在Granger(2005)[1]研究成果的基础上对局部平稳过程的大样本性质进行了深入探讨,发现了一些颇具实际价值的理论结果,弥补了Granger(2005)仅利用预测效果标准得到金融数据生成过程的不足,进一步给出了适合于实证分析的判断金融数据生成过程的标准,并在此基础上详细讨论了局部平稳过程、稳定过程及GARCH模型在大样本情况下的区别。本文运用研究得到的结果,在非平稳框架下对中国股票市场上证180指数进行了分析,发现上证180指数收益率具有明显的非平稳特性,并在此基础上进一步讨论了中国股票市场的市场有效性问题。

关键词: 局部平稳过程, 稳定过程, GARCH模型, 股票市场有效性

Abstract: Based on the research of Granger(2005), we discover the large sample properties of local stationary process to find some valuable theoretical results, which make up the shortage of Granger’s research. In this paper, we put forward the standard of judging financial data generating process and discuss differences in local stationary process, stable process and GARCH model under the situation of large sample. At last, we analyze index 180 of Shanghai security market under the frame of non-stationary, discover that it obviously has non-stationary property and discuss the efficiency of security market in China.

 

Key words: Local stationary process, Stable process, GARCH model, Efficiency of security market