[1]苏梽芳,胡日东,陈家干.中国股市长记忆检验的滑动分块自助法仿真[J].华侨大学学报(自然科学版),2009,30(3):338-342.[doi:10.11830/ISSN.1000-5013.2009.03.0338]
　SU Zhi-fang,HU Ri-dong,CHEN Jia-gan.The Test of Long Memory of China’s Stock Market Based on Moving-Block Bootstrap Simulation[J].Journal of Huaqiao University(Natural Science),2009,30(3):338-342.[doi:10.11830/ISSN.1000-5013.2009.03.0338]
点击复制

中国股市长记忆检验的滑动分块自助法仿真()
分享到：

《华侨大学学报（自然科学版）》[ISSN:1000-5013/CN:35-1079/N]

卷:
第30卷
期数:
2009年第3期
页码:
338-342
栏目:
出版日期:
2009-05-20

文章信息/Info

Title:
The Test of Long Memory of China’s Stock Market Based on Moving-Block Bootstrap Simulation
文章编号:
1000-5013(2009)03-0338-05
作者:
苏梽芳; 胡日东; 陈家干
华侨大学商学院
Author(s):
SU Zhi-fang;  HU Ri-dong;  CHEN Jia-gan
College of Commerce, Huaqiao University, Quanzhou 362021, China
关键词:
长记忆;  分整自回归移动平均模型;  滑动分块自助法;  GPH检验法;  中国股市
Keywords:
long-term memory;  auto-regressive fractiona lintegrated moving average;  moving-block bootstrap approach;  GPH test;  China stock market
分类号:
O212.1
DOI:
10.11830/ISSN.1000-5013.2009.03.0338
文献标志码:
A
摘要:
提出基于滑动分块自助的GPH(Geweke,Porter-Hudak)检验方法,并检验沪、深股市的各种收益率序列的长记忆性.结果表明,沪、深两市的日指数收益率序列是一个I(0)过程而非长记忆过程.然而,沪、深两市的绝对值收益率和收益率平方序列却是一个分数差分过程,沪市的绝对值收益率和收益率平方序列的分数差分值大约为0.30,深市的绝对值收益率和收益率平方序列分数差分值大约为0.35.即深市的长记忆性强于沪市,同时也说明上海股市的运行效率要高于深圳股市.
Abstract:
This paper proposes a test of GPH(Geweke,Porter-Hudak)based on the moving-block bootstrap approach and tests the long-term memory of daily return rate sequence both in Shanghai stock market and Shenzhen stock market,which shows that daily return rate sequence is an I(0) process and not a long-term memory.However,the absolute return rate and squared return rate sequence are fractional differencing process.The fractional differencing parameter of absolute return rate and squared return rate sequence in Shanghai stock market is about 0.35,and the fractional differencing parameter of Shenzhen’s is about 0.30,i.e.the long-term memory of Shenzhen stock market is stranger than Shanghai’s,and it also indicates that Shanghai stock market is more efficient than Shenzhen’s.

参考文献/References:

[1] 陈梦根. 中国股市长期记忆的实证研究 [J]. 经济研究, 2003(3):70-78.
[2] PILAR G C. Tests of long memory:A bootstrap approach [J]. Computational Economics, 2005, (25):103-113.
[3] GEWEKE J, PORTER-HUDAK S. The estimation and application of long memory time series models [J]. Journal of Time Series Analysis, 1983(4):221-238.
[4] 张术林, 魏正红. 金融资产收益非对称性分析 [J]. 深圳大学学报(人文社会科学版), 2007(1):81-84.doi:10.3969/j.issn.1000-260X.2007.01.013.
[5] RICHARD M V. The moving blocks bootstrap versus parametric time series modela [J]. Water Resources Research, 1996(6):1875-1882.
[6] POLITIS D N, WHITE H. Automatic block-length selection for the dependent bootstrap [J]. Econometric Reviews, 2004(1):53-70.
[7] ZHUAN Xin-ding, CLIVE W J G. Modelling volatility persistence of speculative returns:A new approach [J]. Journal of Econometrics, 1996, (73):185-215.

备注/Memo

备注/Memo:
国家软科学研究计划项目(2008GXS5D130); 国家社会科学基金资助项目(08BJL019)
更新日期/Last Update: 2014-03-23