基于MEM模型的金融市场相关性分析与波动溢出研究

发布时间：2018-03-20 04:12

本文选题：乘积误差模型　切入点：高频数据　出处：《天津大学》2012年硕士论文　论文类型：学位论文

【摘要】：随着经济全球化、金融一体化的不断深入,世界各国间的经济、金融联系日益紧密,金融市场波动剧烈,金融市场之间的相关性不断增强,关系日趋复杂。而金融市场波动溢出效应体现了不同金融市场间的波动可能存在的相互影响和金融市场间的危机传染机制。金融市场相关性分析和波动溢出效应的研究已成为金融学术界的研究热点。金融高频数据计量经济学自20世纪90年代以来成为了金融计量学、金融工程学的一个重要研究领域。而高频数据的波动建模是其一项重要的研究方向,其研究将有助于从金融市场微观结构的角度深入分析金融市场的波动特征。乘积误差模型(Multiplicative Error Model,MEM)由Engle(2002)提出,其通过对高频数据下非负(non-negative)金融数据如“已实现”波动(Realized Volatility,RV)建模有效刻画了高频数据的波动特征。本文即是基于MEM模型从高频数据和市场微观结构理论的角度开展对金融市场相关性分析和波动溢出效应的研究。论文的主要研究工作和创新点如下:1、利用“已实现”波动的一种改进方法——调整“已实现”波动作为变量将MEM模型应用于中国股票市场进行实证分析,研究高频数据下中国金融市场的波动建模问题;2、结合MEM模型,分析研究在高频数据下中国上海和深圳股市之间相互之间的波动溢出问题;3、建立适当的Copula-MEM模型,对高频数据下中国上海和深圳股市之间的相关程度和相关模式进行研究;4、将门限模型与Copula函数结合,构建两体制门限MEM模型,分析不同体制下的高频数据波动建模,并与Copula函数相结合,通过观察其不同波动体制下Copula-MEM模型所体现的相关模式和相关程度的差异来分析研究金融市场之间的波动溢出。
[Abstract]:With the economic globalization and the deepening of financial integration, the economic and financial links between countries in the world are increasingly close, the financial market fluctuates violently, and the correlation between the financial markets is increasing. The relationship is becoming more and more complex. The volatility spillover effect of financial market reflects the possible interaction of volatility among different financial markets and the crisis contagion mechanism between financial markets. Correlation Analysis and volatility spillover effect of Financial Market. Financial high-frequency data econometrics has become financial metrology since 1990s. Financial engineering is an important research field, and volatility modeling of high-frequency data is one of the important research directions. Its research will help to analyze the volatility characteristics of financial markets from the perspective of the microstructure of financial markets. The multiplicative Error Model was proposed by Engle2002. By modeling non-negative non-negative) financial data under high frequency data, such as "realized" volatility model, this paper describes the volatility characteristics of high frequency data effectively. This paper is based on MEM model from the perspective of high frequency data and market microstructure theory. The main research work and innovation of this paper are as follows: 1, using "realized" volatility as a kind of improvement method-adjusting "realized" volatility as a variable to use "realized" volatility as a variable. The model is applied to the empirical analysis of Chinese stock market. This paper studies the volatility modeling problem of Chinese financial market under high frequency data. Combined with MEM model, the volatility spillover problem between Shanghai and Shenzhen stock markets in China under high frequency data is analyzed, and an appropriate Copula-MEM model is established. This paper studies the correlation degree and correlation model between Shanghai and Shenzhen stock markets under high frequency data. Combining threshold model with Copula function, the threshold MEM model of two systems is constructed, and the volatility modeling of high frequency data under different systems is analyzed. Combined with the Copula function, the volatility spillover between the financial markets is analyzed by observing the correlation model and the correlation degree of the Copula-MEM model under different volatility systems.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F224;F832.51

【相似文献】