| 中国沪深股市均值回复效应实证研究——基于GIBBS抽样方法 |
| |
| 引用本文： | 胡日东,苏梽芳,李立柱. 中国沪深股市均值回复效应实证研究——基于GIBBS抽样方法[J]. 宁波大学学报(人文科学版), 2008, 21(6): 70-77 |
| |
| 作者姓名： | 胡日东 苏梽芳 李立柱 |
| |
| 作者单位： | 华侨大学,商学院,福建,泉州,362021;华侨大学,商学院,福建,泉州,362021;华侨大学,商学院,福建,泉州,362021 |
| |
| 基金项目： | 国家社会科学基金，福建省自然科学基金 |
| |
| 摘 要： | 使用一种新的模拟方法：GIBBS抽样对沪深股市月超额收益率数据进行标准化后,基于VR检验研究两市月度超额收益率是否存在均值回复效应。实证结果表明,1996年到2006年的11年间,在间隔为3-33个月的水平上,沪深股市都不存在均值回复效应,即两个市场都达到了弱式有效,而沪市的有效性大于深市。传统的抽样方法只考虑数据异方差的持续性而忽略异方差的具体形式,文章基于一个Markov-Switch模型进行GIBBS抽样,在估计VR的经验分布时考虑了数据异方差的持续性和具体形式等信息,因此结论更加可靠。
|
| 关 键 词： | GIBBS抽样 方差比检验 均值回复 |
An Empirical Study of Mean Reversion of Shanghai and Shenzhen Securities Exchanges-Based on GIBBS |
| |
| HU Ri-dong,SU Zhi-fang,LI Li-zhu. An Empirical Study of Mean Reversion of Shanghai and Shenzhen Securities Exchanges-Based on GIBBS[J]. Journal of Ningbo University(Liberal Arts Edition), 2008, 21(6): 70-77 |
| |
| Authors: | HU Ri-dong SU Zhi-fang LI Li-zhu |
| |
| Affiliation: | ( Faculty of Business, Ovemeas Chinese University, Qqanzhou 362021, China ) |
| |
| Abstract: | This paper is based on a test to find out whether there exists mean reversion effect or not with the monthly excess return of Shanghai index and Shenzhen index by employing GIBBS sampling to standardize the return series.The result shows no evidence about the mean reversion in the two stock markets during the period 1996-2006,which indicates that both markets are weak-form efficient.Based on a Markov-Switch model,in estimation of the VR empirical distribution we allow for the information on heteroskedacity such as the persistence and specific pattern,so the conclusion should be more reliable. |
| |
| Keywords: | GIBBS sampling randomization variance ratio mean reversion |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
