中国股市的日内效应、长记忆性及多重分形性:基于价-量交叉相关性视角  

Intraday Effects,Long Memory and Multifractal Characteristics in Chinese Stock Markets:a Perspective from Cross Correlation between Price and Trading Volume

作　　者：苑莹[1] 庄新田[1] 金秀[1] 

机构地区：[1]东北大学工商管理学院,沈阳110819

出　　处：《系统管理学报》2016年第1期28-35,共8页Journal of Systems & Management

基　　金：国家自然科学基金资助项目(71271047;71571041;71371044);教育部新世纪优秀人才支持计划资助项目(NCET-13-0115);中央高校基本科研业务费资助项目(N140604004);辽宁省高等学校优秀人才支持计划资助项目(LJQ2013030);辽宁省自然科学基金资助项目(2015020077)

摘　　要：与以往仅针对价格或交易量某单一序列的实证研究不同,从价-量交叉相关性视角,以金融危机期间中国股市每1分钟高频数据为研究对象,对其日内效应、长记忆性及多重分形性等异象性特征进行了研究。首先,对我国股市的日内效应进行了检验,发现该日内效应表现为"U"型形式。进一步,运用DFA和DCCA方法,研究了中国股市价格和成交量序列的长记忆性及交叉相关性特征。最后,运用MF-DFA和MF-DCCA方法对中国股市价格及交易量序列进行了多重分形分析和多重分形交叉相关分析。上述研究结果表明,中国股市价格序列与交易量序列不但自身具有显著的长记忆性及多重分形特征,且股市价-量关系之间也存在着显著的长记忆性和多重分形特征,即金融市场的价格变动不但受价格变动本身的影响,还会受到交易量变动的影响。这说明,对股市行为的分析应将价、量作为一个有机整体来全面考虑,而不是进行单独量化。Most literature focuses on the analysis of assets price or trading volume but ignores their cross correlation. Aiming at this problem, from the perspective of cross correlation between price and trading volume, an empirical research on intraday effects, long memory and multifractal characteristics is conducted using the high frequency data in the period of financial crisis in the Chinese stock markets. Firstly, the intraday effect in stock markets is tested and confirmed. In addition, a detrended fluctuation analysis and detrended cross correlation analysis are conducted. Finally, a multifractal detrended fluctuation analysis and multifractal detrended cross correlation analysis are conducted as well. It is found that there is not only long memory in each of price series and trading volume series but also long memory between the price and trading volume series. Similarly, it is found that there is not only multifractal in each of price series and trading volume series but also between the price and trading volume series. These indicate that there is a significant cross correlation and interaction between the price and trading volume.Hence we suggest that ＂price＂ and ＂trading volume＂ should be used as a whole to study the Chinese stock market.

关 键 词：高频数据 日内效应 DCCA MF-DCCA 

分 类 号：F830.9[经济管理—金融学]