摘要
对于中国股市的长记忆性问题,许多的学者得到了不同的结论,这主要是由于不同方法自身存在着缺陷,针对这一问题,分别运用经典R/S分析,修正R/S分析,DFA及广义Hurst指数等统计分析方法对其进行实证研究,以期使结果更加稳健.结果发现沪深两市均具有长记忆性特征.此外,研究了广义Hurst指数值与市场发展状态之间的依赖性,并根据该性质得到结论认为沪深两市属于新兴的、非有效的市场.
About long memory of Chinese stock markets, some different conclusions have been drawn. This is because that the methods have their own limitation. Aimed at this issue, an empirical study of stock markets in China was presented by using R/S, modified R/S, DFA and generalized Hurst exponents. It is found that long memory of markets is confirmed. In addition, the correlation of general Hurst exponents and state of market development is researched and it is found that the Chinese markets belong to the category of emerging financial markets.
出处
《数理统计与管理》
CSSCI
北大核心
2008年第1期156-163,共8页
Journal of Applied Statistics and Management
基金
国家自然科学基金(70371062)