基于g-h模型下的VaR估计及应用 Alternative Title Estimate and Applications of VaR Based on the Model of g-h 丁芳 Thesis Advisor 严定琪 2013-05-26 Degree Grantor 兰州大学 Place of Conferral 兰州 Degree Name 硕士 Keyword VaR g-h分布 极值理论 Abstract 许多金融机构由于管理不善而导致巨额损失，其主要原因是缺乏有效的风险管理工具，从而不能很好的评估市场风险. 因此，进行有效的市场风险评估就显得尤其重要. 市场风险的度量工具有许多，VaR是目前金融市场风险管理常用的方法之一，它表示某个金融资产或投资组合在一定持有期和置信水平下的最大可能损失. 我们可以根据证券组合价值变化的统计分布图找到与置信度匹配的分位数，即VaR值.通常假定金融资产或投资组合的收益率序列服从正态分布，但是大量实证表明资产收益率是非正态的，而且具有明显的尖峰厚尾特性，极值理论(EVT)能较好的处理尖峰厚尾特性，但这一方法有很多的缺陷，计算方法非常复杂，阀值的选择比较自由，而且这一方法对样本数据容量要求比较大. 针对极值理论存在的这些不足，本文采用g−h分布，导出了基于g−h分布下投资组合的VaR计算公式，并且与极值理论下的VaR估计值做了比较，实证分析结果表明基于g−h假设下的VaR方法对收益率数据的风险描述更为准确，计算结果优于极值理论下的VaR估计. Other Abstract Many financial institutions generate huge loss because of mismanagement, the main reason is lack of effective risk measurable tool, so market risk can not be assessed well.It is visible that effective assessment of market risk is particularly important. There are several market risk measurement tools, VaR is one of popular methods of financial market risk management, it denotes maximal possible loss of a financial asset or portfolio in a certain holding period and under the confident level.We can find fraction matching with confidence according to statistical distribution of the changes of portfolio value, i.e VaR. Generally, we suppose that interests series of the financial asset or portfolio is normal distribution, but lots of empirical evidences show that return rate of asset is non-normal and has obvious peak fat-tail features. Extreme value theory(EVT) can handle peak fat-tailed features well, but this method has many defects, such as complicated calculations and subjectivity of the threshold's selection. For these deficiencies of extreme value theory, we use the g-h distribution to derive the VaR calculation formula of portfolio based on the g-h distribution in this paper and compare with extreme value theory. By empirical analysis, results show that VaR method based on g-h more accurately describe the risk of interests series and that the calculation results is better than the VaR estimation based on the extreme value theory. URL 查看原文 Language 中文 Document Type 学位论文 Identifier https://ir.lzu.edu.cn/handle/262010/225230 Collection 数学与统计学院 Recommended CitationGB/T 7714 丁芳. 基于g-h模型下的VaR估计及应用[D]. 兰州. 兰州大学,2013.
 Files in This Item: There are no files associated with this item.
 Related Services Recommend this item Bookmark Usage statistics Export to Endnote Altmetrics Score Google Scholar Similar articles in Google Scholar [丁芳]'s Articles Baidu academic Similar articles in Baidu academic [丁芳]'s Articles Bing Scholar Similar articles in Bing Scholar [丁芳]'s Articles Terms of Use No data! Social Bookmark/Share
No comment.