沪深港股市间的非对称相依结构和风险传导——去除共同影响因素的视角  

Asymmetric interdependence structure and risk transmission between Shanghai,Shenzhen and Hong Kong stock markets with the common influencing factors removed

作　　者：曹广喜 谢文浩 CAO Guang-xi;XIE Wen-hao(School of Management Science and Engineering,Nanjing University of Information Science&Technology,Nanjing 210044,China)

机构地区：[1]南京信息工程大学管理工程学院,南京210044

出　　处：《管理科学学报》2025年第3期148-161,共14页Journal of Management Sciences in China

基　　金：国家社会科学基金资助重大项目(19ZDA105);江苏省高校哲学社会科学研究项目(2022SJZD018);江苏省研究生科研创新项目(KYCX22_1242).

摘　　要：由于多重分形去趋势偏相关分析方法(MF-DPXA)无法测度不同趋势(上涨和下跌)下非对称相依关系问题,本文提出多重分形非对称去趋势偏相关分析方法(MF-ADPXA);进一步提出去因素时滞去趋势交叉相关分析法(ETD-DCCA)研究股市间的风险传导方向.以上证指数、深成指数和恒生指数为研究对象,实证分析去除共同影响因素后两两股市间非对称交叉相关性和风险传导.结果表明,去除某一股市的影响后,其余两股市间长记忆交叉相关性较弱,收益率上涨时长记忆交叉相关性增强,收益率下跌时交叉相关性表现为反持续性;大幅波动时的非对称程度增强.两两股市间的局部交叉相关性随着时间推移具有减弱趋势.随着时滞的增加,两两股市间反持续交叉相关性有所增强,深成指数风险主要传导至上证指数,上证指数风险主要传导至恒生指数.恒生指数主要对深成指数影响更强.上述实证结论对于投资者重新认识沪深港股市本质相依结构和风险传导,合理进行跨市场投资组合以及监管者防范系统性金融风险具有重要的启示意义.Since the multifractal detrended partial correlation analysis method(MF-DPXA)cannot measure the asymmetric dependence relationship under different trends(upward and downward),this paper proposes the multifractal asymmetric detrended partial cross-correlation analysis method(MF-ADPXA).Furthermore,the paper proposes a removing factors time-delayed detrended cross-corelation analysis(ETD-DCCA)to study the risk transmission direction between stock markets.Taking Shanghai Component Index,Shenzhen Component Index,and Hang Seng Index as research objects,this paper empirically analyzes the asymmetric crosscorrelation and risk transmission between pairwise stock markets after removing the common influencing factors.The results show that,after removing the influence of one stock market,the long-memory cross-correlation between the other two stock markets is weak.When the return trend is upward,the long memory crosscorrelation increases,and when the return trend is down,the cross-correlation shows anti-persistence.The degree of asymmetry is greater when the fluctuation is large.The local cross correlation between the pairwise stock markets shows a weakening trend over time.As the time lag increases,the anti-persistent cross-correlation between the two-two stock markets is enhanced.The risk of the Shenzhen Component index is mainly transmitted to the Shanghai Component Index,and the risk of the Shanghai Component Index is mainly transmitted to the Hang Seng Index.The Hang Seng Index has a stronger impact on the Shenzhen Component Index.This study has implications for re-understanding the intrinsic dependent structure and risk transmission of Shanghai,Shenzhen,and Hong Kong stock markets,cross-market porfolio,and risk management.

关 键 词：多重分形 去趋势交叉偏相关性 非对称性 影响因素 

分 类 号：F832.5[经济管理—金融学]