基于波动约束性度量的中美港股市交叉相关性研究
发布时间：2018-11-16 18:29
【摘要】：本文运用多重分形消除趋势交叉相关性析(MF-DCCA)法,在不同波动约束区间上对中国大陆,美国和香港股市进行了研究。本文提出了基于波动约束性度量的多重分形消除趋势交叉相关性分析法(VC-MF-DCCA)研究中国大陆,美国和香港股市间的波动传导。实证分析结果表明,股市波动与金融市场重要事件相关。以恒生指数(HSI)为样本数据的香港股市的影响比以上证指数为样本数据的中国大陆股市大。而以上证指数为样本数据的中国大陆股市比以道琼斯工业平均指数(DJIA)为样本数据的美国股市影响大。在香港股市和中国大陆股市间的传导性最强。在1991年至2014年间的大幅波动区间内,香港股市影响最大。本文采用了分整自回归移动平均法验证了基于波动约束性度量的多重分形消除趋势交叉相关性分析法(VC-MF-DCCA)的有效性。本文将基于波动约束性度量的多重分形消除趋势交叉相关性分析法(VC-MF-DCCA)与基于波动约束性度量的消除趋势交叉相关性分析(VC-DCCA)法以及基于经验模态分解(EMD)的多重分形消除趋势交叉相关性分析法比较,研究发现基于波动约束性度量的多重分形消除趋势交叉相关性分析法比较精确,也能直观地与具体的股市大事件相联系。
[Abstract]:In this paper, the multi-fractal elimination trend cross-correlation analysis (MF-DCCA) method is used to study the stock markets in mainland China, the United States and Hong Kong in different volatility constraints. In this paper, a multi-fractal elimination trend cross-correlation analysis (VC-MF-DCCA) based on volatility constraint metric is proposed to study the volatility conduction between the mainland of China, the United States and Hong Kong stock markets. The empirical results show that stock market volatility is related to important events in financial markets. Hong Kong's stock market, based on the Hang Seng index (HSI), is more influential than the mainland stock market, which is based on the Shanghai index. Mainland Chinese stocks with the Shanghai index as a sample are more influential than U.S. stocks with the Dow Jones Industrial average (DJIA) as a sample. The Hong Kong stock market and the mainland Chinese stock market are the most conductive. Hong Kong's stock market was the most influential in the sharp range of volatility between 1991 and 2014. In this paper, the split autoregressive moving average method is used to verify the validity of multifractal trend elimination cross correlation analysis (VC-MF-DCCA) based on volatility constraint measure. In this paper, the multifractal elimination trend cross correlation analysis (VC-MF-DCCA) based on volatility constraint metric, the cross correlation analysis (VC-DCCA) method based on volatility constraint metric and the empirical analysis method are presented. The cross-correlation analysis method of multifractal elimination trend based on modal decomposition (EMD) is compared. It is found that the multi-fractal elimination trend cross-correlation analysis method based on volatility constraint measure is more accurate and can be intuitively associated with specific major events in the stock market.
【学位授予单位】：南京信息工程大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F831.51
本文编号：2336293
[Abstract]:In this paper, the multi-fractal elimination trend cross-correlation analysis (MF-DCCA) method is used to study the stock markets in mainland China, the United States and Hong Kong in different volatility constraints. In this paper, a multi-fractal elimination trend cross-correlation analysis (VC-MF-DCCA) based on volatility constraint metric is proposed to study the volatility conduction between the mainland of China, the United States and Hong Kong stock markets. The empirical results show that stock market volatility is related to important events in financial markets. Hong Kong's stock market, based on the Hang Seng index (HSI), is more influential than the mainland stock market, which is based on the Shanghai index. Mainland Chinese stocks with the Shanghai index as a sample are more influential than U.S. stocks with the Dow Jones Industrial average (DJIA) as a sample. The Hong Kong stock market and the mainland Chinese stock market are the most conductive. Hong Kong's stock market was the most influential in the sharp range of volatility between 1991 and 2014. In this paper, the split autoregressive moving average method is used to verify the validity of multifractal trend elimination cross correlation analysis (VC-MF-DCCA) based on volatility constraint measure. In this paper, the multifractal elimination trend cross correlation analysis (VC-MF-DCCA) based on volatility constraint metric, the cross correlation analysis (VC-DCCA) method based on volatility constraint metric and the empirical analysis method are presented. The cross-correlation analysis method of multifractal elimination trend based on modal decomposition (EMD) is compared. It is found that the multi-fractal elimination trend cross-correlation analysis method based on volatility constraint measure is more accurate and can be intuitively associated with specific major events in the stock market.
【学位授予单位】：南京信息工程大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F831.51
【参考文献】
相关期刊论文 前10条
1 谢家泉;;中美股市风险传染效应研究综述——基于金融危机视角[J];韶关学院学报;2016年11期
2 戚琦;汪凯;吴齐;;基于GARCH族模型的深证成分指数波动性研究[J];科技和产业;2015年09期
3 陆剑清;陈明珠;;投资者情绪:影响股市波动的无形之手[J];金融市场研究;2015年09期
4 苏跃辉;陈扬;;基于ARCH族模型的沪深股市波动性及动态相关性研究[J];金融教学与研究;2015年03期
5 龚金国;史代敏;;金融自由化、贸易强度与股市联动——来自中美市场的证据[J];国际金融研究;2015年06期
6 蔡庆丰;郭俊峰;陈耀辉;;沪深300期现货市场动态波动关系研究:基于VECM-GJR-DCC-MGARCH-t模型的视角[J];系统管理学报;2015年02期
7 唐勇;陈艳茹;;考虑杠杆效应的多重分形波动建模:基于中国股指的实证分析[J];系统工程学报;2015年01期
8 高大良;刘志峰;杨晓光;;投资者情绪、平均相关性与股市收益[J];中国管理科学;2015年02期
9 冯玉梅;陈璇;张玲;;融资融券交易对我国股市波动性的影响研究——基于融资融券转常规前后的比较检验[J];山东社会科学;2015年02期
10 陈耀辉;郭俊峰;殷文超;;人民币升值对中小板市场波动的影响——基于贝叶斯分位数回归的分析[J];系统工程;2015年01期
，本文编号：2336293
本文链接：https://www.wllwen.com/jingjilunwen/touziyanjiulunwen/2336293.html
