基于Copula-VaR方法的沪深股市投资组合风险分析

发布时间：2018-03-07 19:16

本文选题：Copula-VaR　切入点：金融风险管理　出处：《华中科技大学》2013年硕士论文　论文类型：学位论文

【摘要】：近年来，伴随着信息技术的发展，全球化进程的脚步也越来越快，经济全球化使得各个国家和各个市场之间的联系更为紧密，单一市场波动所带来的影响能够轻而易举地跨越市场、跨越国界，对处在全球一体化下的其他国家和市场产生不同程度的影响。而股票市场作为支持一国经济发展并控制国家经济命脉的最重要的资本市场，其重要性不言而喻。因此，研究股票市场的风险并以此来防范其对国家经济建设可能带来的不利影响就显得尤为重要。理论部分首先介绍了金融风险的定义和分类，其中VaR因其容易理解、易于操作且能够量化风险等诸多特点现已成为风险监管机构主流的风险测度方法，即便VaR方法有着这些优点，但与此同时其也有不足的地方，于是在此基础上提出了Copula-VaR方法来对股票市场进行风险分析。其次对Copula函数的性质和不同类型Copula函数的特征进行了详细的介绍，因为Copula在测度资产组合的风险时是不需要资产收益率服从正态分布的假设前提，而且其在结构上能较好的拟合资产之间的联合分布，并且也能消除单纯的VaR因不能解决资产之间的非线性相关性所引起的不合理测度风险等特点，因此Copula-VaR方法能得到更加精确的投资组合VaR值。实证部分选择我国股票市场的上海综合指数和深圳成分指数进行研究分析。根据样本数据联合密度函数所具有的特征，本文选用的是具有尾部对称特征的二元t-Copula函数作为上证指数和深成指数的联合分布函数，利用Copula函数对其各自变量的求偏导后服从0-1均匀分布的特点，用蒙特卡洛模拟法求出了在不同置信水平下的VaR。最后，变动投资组合中上证指数和深成指数的比例就得到了一组VaR值并将得到的结果绘制成曲线，随着模拟次数的增加，得到了近似直线的投资比例和组合的VaR值，直线的斜率即是单位上证指数百分比变动所引起的组合VaR值变化，也就是全文的结论。文章中所用到的Copula方法是最近发展起来的一种更加精确的用来测度相关性的方法，不仅适用于股票市场，对于其他资本市场同样适用，，另外，这种方法还可以解决多变量相关性的问题。
[Abstract]:In recent years, with the development of information technology, the pace of globalization has become faster and faster. Economic globalization has made countries and markets more closely linked. The impact of single market volatility can easily cross markets, across borders, The importance of the stock market as the most important capital market to support a country's economic development and control the lifeblood of a country's economy is self-evident. It is very important to study the risk of stock market and prevent its possible adverse effects on national economic construction. The theoretical part first introduces the definition and classification of financial risk. VaR has become the mainstream risk measurement method for risk regulators because of its easy understanding, easy operation and the ability to quantify risk. Even if the VaR method has these advantages, at the same time it has its shortcomings, On this basis, Copula-VaR method is proposed to analyze the risk of stock market. Secondly, the properties of Copula functions and the characteristics of different types of Copula functions are introduced in detail. Because Copula does not need to assume the return of assets from normal distribution when measuring the risk of portfolio, and it can fit the joint distribution of assets better in structure. And it can also eliminate the characteristics of VaR which can not solve the nonlinear correlation between assets caused by the unreasonable risk measurement, so Copula-VaR method can get more accurate portfolio VaR value. In the empirical part, the Shanghai Composite Index and the Shenzhen component Index of China's stock market are selected for research and analysis. According to the characteristics of the sample data associated with the density function, In this paper, the binary t-Copula function with tail symmetry is chosen as the joint distribution function of Shanghai Stock Exchange Index and Deep Index, and the Copula function is used to obtain the partial derivative of their variables from 0-1 uniform distribution. At last, the ratio of Shanghai Stock Exchange Index to Deep Index in the variable portfolio is obtained by Monte Carlo simulation method. Finally, a set of VaR values are obtained and the obtained results are drawn into curves. With the increase of simulation times, The investment ratio and the VaR value of the portfolio are obtained. The slope of the straight line is the change of the VaR value caused by the percentage change of the stock index, which is the conclusion of the whole paper. The Copula method used in this paper is a more accurate method for measuring correlation, which is developed recently. It is applicable not only to the stock market, but also to other capital markets. This method can also solve the problem of multivariate correlation.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F224;F832.51

【相似文献】