基于极值-Copula模型的中国金融市场系统风险的溢出效应研究

发布时间：2018-03-02 18:10

  本文选题：系统风险　切入点：风险溢出效应　出处：《吉林大学》2017年硕士论文　论文类型：学位论文

【摘要】：在全球金融体系逐渐趋于一体化的今天,系统性金融风险的爆发频率有所增加,因此对系统性金融风险的研究一直是经济学者们探讨的重点。随着度量金融风险理论的不断提出,对系统性风险的研究已经从整体细分到区域。本文在此背景下,测量当突发事件发生时,银行、证券、保险三个金融市场之间风险溢出效应变化情况。在以往的研究成果中,对于系统性金融风险的度量经历了定性到定量的分析过程。提出了Va R(在险价值)概念来度量金融风险大小。在之后的发展中该方法被逐渐完善,相继出现了CAVia R方法以及Co Va R方法,其中Co代表着条件性和传染性,从度量模型的发展过程也可以看出系统性风险的研究从单个市场转到了市场之间的风险联动效应上。但在对于金融时间序列数据的处理和刻画方面,模型的选择仍然有待完善。本文选择构建EGARCH-POT-Copula模型,来测量三个市场间的Co Va R值(联动Va R),即某一市场对其他金融市场的风险联动值。模型运用中发现,EGARCH模型对刻画收益率序列的非对称性具有良好的拟合效果,同时对于残差项的尖峰厚尾非正态性我们采用极值理论中的POT模型进行拟合,效果良好,从而得出三个子市场各自的Va R值。最后,引入数学领域中可以灵活表达非线性、非对称关系的Clayton Copula函数来测量各市场间的Co Va R值。本文共分为五个部分,在第一部分绪论中,简要阐述了选题的背景及意义,概述了系统性金融风险以及EGARCH-POT-Copula模型的发展历史。在第二部分本文从理论分析角度分别阐述了三个市场系统性风险产生的原因,共同的因素包括政策制度不完善、金融市场不发达等。对市场间风险传导机制进行了理论分析,包括直接传导机制的三种渠道:融资风险渠道、支付环节、资产负债渠道,以及间接传导因素羊群效应和市场间业务趋同现象。第三部分构建EGARCH-POT-Copula模型。第四部分选取了2007年至2017年申银万国二级行业数据进行实证研究得出的结果表明,第一,银行业对证券业的溢出风险值最大,表明银行业发生突发事件时对证券业的影响最大,并且证券业对银行的溢出风险值也明显高于对保险业,说明银行证券两个子市场之间的风险联动效应最强。第二,我们注意到虽然目前保险业对其他两个市场的溢出风险值远远小于银行业,但是随着我国保险市场规模的扩大,保险业资本金逐渐开始在其它金融子市场中活跃,因此我们提出对保险业溢出风险也要进行重点观测,严加防范。第五部分本文根据实证分析的结论给出关于监管市场间风险溢出效应的政策建议:基于风险溢出效应的存在性,稳定金融市场的前提是要对风险传导渠道进行关注。当某一市场爆发系统性风险危机时其他金融子市场应做出迅速应对措施,最大程度降低风险的传染。在预防危机方面我们要重点跟踪金融系统中重要商业银行的系统性风险,严防银行出现极端情况后向其他市场传导风险。同时要循序渐进地推行混业经营,加强宏观审慎监管,避免因风险溢出效应加大引起的系统性金融危机。综上所述,本文运用极值理论中的POT模型与Copula函数结合针对我国金融子市场之间的系统性风险溢出效应进行分析,得出了以下结论:当某一市场爆发系统性风险危机时对其他金融市场会产生冲击,风险的外溢会使危机在短时间内迅速蔓延。加强金融系统性风险危机的监控与预防是金融市场稳定发展的重要条件。
[Abstract]:In the global financial system gradually integration today, increased systemic financial risk outbreak frequency, so the research on the system of financial risk has been the focus of economic scholars. With the financial risk measurement theory being put forward, the study of systemic risk to the region as a whole has been subdivided. In this context, measurement when emergencies occur, banking, securities, insurance between the three financial market risk spillover effect changes. In previous research, the systemic financial risk measurement through the analysis of qualitative and quantitative is proposed. The Va R (value at risk) to measure the financial risk the size. After the development is gradually improved, there have been CAVia R Co Va method and R method, in which Co represents conditional and contagious, from the development process measurement model can also be The research of systematic risk transfer from single market risk linkage effect between the market. But in the processing and characterization of financial time series data, model selection is still to be improved. This paper chooses to build EGARCH-POT-Copula model, to measure the market between the three Co Va R (Va R, the linkage) other financial market linkage risk of a market value. That model application, EGARCH model has good fitting effect on depicting the asymmetry of the return series, while the peak thick tail of residuals from non normality we adopt POT model of extreme value theory in the fitting effect is good, so that the three sub markets the Va value of R. Finally, the flexible nonlinear expression can be introduced in the field of mathematics, asymmetric relationship between the Clayton Copula function to measure the market between the Co Va R value. This paper is divided into five parts, In the first part of the introduction, briefly describes the background and significance of the topic, an overview of the system of financial risk and EGARCH-POT-Copula model of development history. In the second part of this paper illustrates the three market risk causes, common factors include the policy system is not perfect, the underdevelopment of financial markets. The market risk conduction mechanism is analyzed, three kinds of channels, including direct transmission mechanism: financing channels, payment links, asset liability channel, and the indirect conduction factors of herding and market business convergence. The third part is the construction of the EGARCH-POT-Copula model. The fourth part is from 2007 to 2017 two SW level industry data the empirical research results show that the first, spillover risks of the banking industry on the securities industry, the largest value, suggest that the banking industry had burst Impact on the securities industry's biggest event, and the securities industry to overflow bank's risk value is significantly higher than that of the insurance industry, the strongest risk linkage effect between the two sub bank securities market. Second, we note that although spillover risk in the insurance industry at present on the other two market is far less than the value of the banking industry, but with the expansion of the scale of China's insurance market, capital insurance industry gradually became active in other financial markets, so we put forward to the insurance industry spillover risks should focus on observation, to guard against. The fifth part of this paper is given according to the conclusions of Empirical Analysis on the Risk Spillover Effect between market regulatory policy recommendations: the existence of the Risk Spillover Effect Based on the premise of the stability of financial markets is to focus on the risk conduction channels. When a market systemic risk crisis when other financial sub market should be Make rapid measures to minimize the risk of infection in the prevention of the crisis. We should focus on tracking the systemic risk of commercial banks in the financial system, the bank to prevent the extreme situation to other market transmission risk. At the same time to gradually carry out the mixed operation, strengthen macro prudential supervision, to avoid the risk of spillover effects increase system caused by the financial crisis. In summary, this paper uses POT model and Copula function in extreme value theory to analyze the combination between China's financial market system risk spillover effect, draws the following conclusions: when a market systemic risk crisis will have an impact on other financial markets, Risk Spillover will make the crisis spread rapidly in a short time. To strengthen the monitoring and prevention of the risk of financial crisis system is the important conditions for the stable development of the financial market.

【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F224;F832.5

【相似文献】

中国期刊全文数据库 前10条

1 韩明;Copula——一个新的计量经济工具[J];统计与信息论坛;2004年05期

2 李健伦,方兆本,鲁炜,李红星;Copula方法与相依违约研究[J];运筹与管理;2005年03期

3 单国莉,陈东峰;一种确定最优Copula的方法及应用[J];山东大学学报(理学版);2005年04期

4 罗俊鹏;;基于Copula的金融市场的相关结构分析[J];统计与决策;2006年16期

5 李霞;曾霞;侯兵;;copula的构造以及copula之间关系的研究[J];商丘师范学院学报;2006年05期

6 孙志宾;;混合Copula模型在中国股市的应用[J];数学的实践与认识;2007年20期

7 李娟;戴洪德;刘全辉;;几种Copula函数在沪深股市相关性建模中的应用[J];数学的实践与认识;2007年24期

8 许建国;杜子平;;非参数Bernstein Copula理论及其相关性研究[J];工业技术经济;2009年04期

9 杜子平;闫鹏;张勇;;基于“藤”结构的高维动态Copula的构建[J];数学的实践与认识;2009年10期

10 王s，

本文编号：1557610