【作者】 刘军；

【导师】 赖晓东；

【作者基本信息】 西南交通大学 ， 金融学， 2013， 硕士

【摘要】 当前的金融市场,随着金融改革的深入化,金融市场的联系日益紧密,投资组合、金融风险管理更是成为学者们关注的热点问题。投资组合风险的研究,首先就是要研究金融变量之间的相关性关系。对于研究投资组合的相关关系,传统投资组合一般都是以正态分布为假设前提的,但实际金融市场的数据并不都是服从正态分布。同时,相关性分析的常用方法就是线性相关系数的分析,线性相关系数要求变量之间的关系是线性的,而金融资产组合之间的关系并不都是线性相关关系,也有很多资产组合之间存在非线性关系,此时也就无法用线性相关系数来准确地描述金融资产之间的相关性。所以,在正态分布假设下,仅使用线性相关来分析资产组合之间的相关性,进而计算其投资组合的风险价值往往与实际的风险价值是存在差异的。因此本文引入Copula函数,作为分析变量之间相关性的工具,无论是变量之间的线性关系还是非线性关系,Copula函数都能够很好地处理并且较好地描述。此外,目前采用Copula函数研究的相关关系大都假定是不变的,但在实际的金融市场中,国家政策变动、金融事件的发生,都会引起经济市场中各种变量之间各种关系的变动,其中变量之间的相关性就不会静止不变,而是会会随着时间变化而变化,所以,本文在利用Copula函数的基础上又进一步利用时变Copula函数来研究资产组合的相关关系。本文首先是介绍本文的研究背景以及意义,通过回顾国内外研究现状,指出传统关于相关性研究方法的不足之处,并指出Copula理论在分析变量之间相关关系的必要性,同时对已有研究文献进行了评述；然后详尽地介绍Copula函数理论,包括Copula函数的性质、定理、类型,然后介绍了Copula模型的构建方法以及参数估计方法、模型评价和检验方法,最后着重介绍时变t-Copula模型；同时对于VaR也作了全面的阐述,其中介绍了其产生、影响因素以及VaR的三种计算方法,并且举例说明了VaR的计算方法。在介绍了Copula和VaR理论的基础上,将时变Copula函数理论应用到风险管理中的实证分析,通过选择沪深股票市场下的投资组合,先用历史数据确定边缘分布,然后选择几种不同的Copula函数和时变Copula来描述资产之间的相关性,进而度量出相应投资组合风险,并进行比较和评价,发现在时变Copula能够更好地估算资产组合的VaR。更多还原

【Abstract】 With the deepening of financial reform,the relationship of the current financial market is increasing.Portfolio investment and financial risk management have become a hot issue for scholars. As for the study of portfolio risk,First of all,It’s should study the relationship between the correlation between financial variables. The traditional methods were based on normal distribution assumption when researching the correlation of investment portfolio.However,the actual financial market data are not subject to normal distribution.Meanwhile,The coefficient of linear correation is the common methods of correlation analysis,But the relationship between the portfolio of financial assets are not all linear correlation,and some correlations of portfolio are not nonlinear relationship.So it’s unable to describe the correlation of the financial assets with linear correlation now.It exits the difference to value the risk if only using the method of linear correalation.As a result,this paper introduced the Copula function, not only do the Copula function describe the linear correalation between financial assets,but also the nonlinear correalation between financial assets.The relationship between financial assets is assumed constant by using current Copula function.On the present financial market,the national policy change or financial events will cause the changes of correlation between financial assets.So the correlation is changing with the time.The paper will use time-varying Copula function based on the Copula function.This paper introduces the background of the study and the significance.It’s pointed out the shortcomings of traditional methods of correlation and the necessity of Copula theory in the analysis of relationships between variables.Then introduces the mothods,the evalution,test and the time-varying of Copula function.Meanwhile,this paper introduces the emergence,development,three elements and calculating of VaR.Finally, the time-varying Copula function theory is applied to the empirical analysis of risk management by selecting the Shanghai and Shenzheng stock market portfolio.The first is determining the marginal distribution by using historical data.Then measure the corresponding portfolio risk with different Copula functions that are used to describe the relationship of financial assets.Finally,it founds the time-varying Copula function can be better estiamate the portfolio VaR.更多还原