CDO市场中的评级模型套利

发布时间：2018-02-02 07:55

本文关键词： CDO 评级选购 K-S检验判别分析　出处：《复旦大学》2013年硕士论文　论文类型：学位论文

【摘要】：2007年起的金融危机仍在对我们的生活产生巨大的影响。危机初步显现为次级住房抵押贷款的直接违约损失,但却被债务抵押债券(Collateralized Debt Obligations, CDO)放大,许多潜在问题至今仍未爆发出来。CDO类属于结构性金融产品,相较于一般的债券更依赖于独立第三方评级机构对其的评级。而由于在评级问题上,发行人与投资者存在着很强的信息不对称,故导致了评级模型套利(或称评级选购)的出现。本文的思路来源于Fender and Kiff(2005),基于实证数据和统计方法,对基于信息不对称的评级模型套利是否存在进行了统计检验。在单变量检验中使用了单因素ANOVA和K-S检验,在此结果基础上作判别分析,得出多变量检验结果。结论是,基于信息不对称的评级模型套利的确存在,并且CDO的发行币种和期限对其差异的影响十分显著。
[Abstract]:The financial crisis that began in 2007 is still having a huge impact on our lives. The crisis has initially emerged as a direct default loss on subprime mortgages. But they are magnified by collateralized Debt obligations (CDOs). Many potential problems have yet to erupt. The CDO category is a structured financial product, relying more on independent third-party rating agencies than on bonds in general. There is a strong information asymmetry between issuers and investors. This leads to the emergence of rating model arbitrage (or rating selection). The idea of this paper comes from Fender and Kiffon 2005, based on empirical data and statistical methods. This paper makes a statistical test on whether arbitrage exists in the rating model based on information asymmetry. The univariate ANOVA and K-S tests are used in the univariate test, and the discriminant analysis is made on the basis of the results. The results of multivariate test show that arbitrage based on asymmetric information rating model does exist and that the currency and maturity of CDO issue have a significant effect on the difference.
【学位授予单位】：复旦大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F830.9;F224

【参考文献】