基于变结构Copula模型的国内外股市波动溢出效应研究

发布时间：2018-05-08 11:05

本文选题：变结构Copula + Granger因果检验　；参考：《华南理工大学》2013年硕士论文

【摘要】：在经济全球化及金融市场一体化的背景下，金融危机的爆发加剧了风险在全球金融市场之间的传递，这种风险在不同金融市场之间传递的现象即表现为金融市场间的波动溢出效应。虽然目前已有不少关于波动溢出效应的研究，但仍然存在诸多缺陷。因此，本文在对传统波动溢出效应研究方法进行梳理的基础上，提出了一种新的研究方法——EMD分解技术与变结构Copula模型相结合。并通过实证分析验证了该研究方法应用于研究不同股票市场之间的波动溢出效应的有效性。本文的核心工作可概括如下：首先采用EMD技术分解并合成了具有不同经济含义的股票收益率新序列；然后，结合Granger因果检验方法分别对不同地区股票市场的高低频序列进行波动传递的方向分析；最后，，以股票收益率的高频序列作为波动溢出效应的研究对象，建立变结构Copula模型，通过检验Copula函数的相关系数研究分析了发生波动溢出时的强度大小。基于上述方法，以2006年1月4日—2010年12月30日期间的道-琼斯指数（DJI）、英国金融时报指数（FTSE）、日经225指数（NIK）、恒生指数（HSI）、沪深300指数（HS300）每日的收盘价为原始数据，对不同股票市场之间的波动溢出效应进行了深入的分析。研究结果表明：（1）除了日本与中国大陆地区的股票市场，其余股市间均存在一定程度的波动溢出。其中，英美两国股市、日本与中国香港地区股市之间均存在长期的波动溢出效应，并且美国股票市场领跑全球金融市场；（2）英美两国股市由短期内信息双向流动转变为单向信息流动，日本股市与中国股市之间由最初不存在显著的信息流动转变为双向信息流动，其余地区的股市随着交易周期的加长，由单向信息流动转为双向信息流动。发达国家的股价变化往往领先于发展中国家的股价变动。
[Abstract]:In the context of economic globalization and financial market integration, the outbreak of the financial crisis has intensified the transmission of risk among the global financial markets. The phenomenon of risk transmission between different financial markets is the volatility spillover effect between financial markets. Although there are many researches on volatility spillover effect, there are still many defects. Therefore, on the basis of combing the traditional research methods of volatility spillover effect, a new research method, EMD decomposition technique and variable structure Copula model, is proposed in this paper. The validity of this method is verified by empirical analysis in the study of volatility spillover effects between different stock markets. The core work of this paper can be summarized as follows: firstly, we use EMD technology to decompose and synthesize a new series of stock returns with different economic meanings. Combined with Granger causality test method, the direction of volatility transmission of high and low frequency series of stock market in different regions is analyzed respectively. Finally, the variable structure Copula model is established by taking the high frequency series of stock yield as the research object of volatility spillover effect. By testing the correlation coefficient of Copula function, the intensity of volatility spillover is analyzed. Based on the above method, the daily closing prices of the Dow Jones Index (DJI), the Financial Times Index (FTSE), the Nikkei 225 Index (Nikkei), the Hang Seng Index (HSI) and the Shanghai and Shenzhen Index (HS300) for the period from 4 January 2006 to 30 December 2010 are taken as the original data. The volatility spillover effect between different stock markets is analyzed. The results show that there is a certain degree of volatility spillover between the stock markets of Japan and mainland China. Among them, there is a long-term volatility spillover effect between the stock markets of the United States and the United States, the stock markets of Japan and Hong Kong of China. Moreover, the US stock market leads the global financial market.) the stock markets of the United States and the United States have changed from two-way information flow to one-way information flow in the short term, and the information flow between the Japanese stock market and the Chinese stock market has changed from no significant information flow at first to a two-way information flow. The stock market in other regions changed from one-way information flow to two-way information flow with the increase of trading cycle. Stock prices in developed countries tend to move ahead of those in developing countries.
【学位授予单位】：华南理工大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F831.5;F224

【参考文献】