This paper considers the pricing problem of collateralized debt obligations tranches under a structural jump-diffusion model, where the asset value of each reference entity is generated by a geometric Brownian motion ...This paper considers the pricing problem of collateralized debt obligations tranches under a structural jump-diffusion model, where the asset value of each reference entity is generated by a geometric Brownian motion and jump with an asymmetric double exponential distribution. Conditioned on the common factor of individual entity, this paper gets the conditional distribution, and further obtains the loss distribution of the whole reference portfolio. Based on the semi-analytic approach, the fair spreads of collateralized debt obligations tranches, i.e., the prices of collateralized debt obligations tranches, are derived.展开更多
This paper aims to reveal the mechanism of Collateralized Debt Obligations (CDOs) and how CDOs extend the current global financial crisis. We first introduce the concept of CDOs and give a brief account of the de-velo...This paper aims to reveal the mechanism of Collateralized Debt Obligations (CDOs) and how CDOs extend the current global financial crisis. We first introduce the concept of CDOs and give a brief account of the de-velopment of CDOs. We then explicate the mechanism of CDOs within a concrete example with mortgage deals and we outline the evolution of the current financial crisis. Based on our overview of pricing CDOs in various existing random models, we propose an idea of modeling the random phenomenon with the feature of heavy tail dependence for possible implements towards a new random modeling for CDOs.展开更多
Copula method has been widely applied to model the correlation among underlying assets in financial market. In this paper, we propose to use the multivariate Frechet copula family presented in J. P. Yang et al. [Insur...Copula method has been widely applied to model the correlation among underlying assets in financial market. In this paper, we propose to use the multivariate Frechet copula family presented in J. P. Yang et al. [Insurance Math. Econom., 2009, 45：139 147] to price multivariate financial instruments whose payoffs depend on the k^th realization of the underlying assets and collateralized debt obligation （CDO）. The advantage of the multivariate Frechet copula is discussed. Empirical study shows that such copula family gives a better fitting to CDO＇s market price than Gaussian copula for some derivatives.展开更多
基金Supported by the National Natural Science Foundation of China (70771018)the Natural Science Foundation of Shandong Province (2009ZRB019AV)Mathematical Subject Construction Funds and the Key Laboratory of Financial Information Engineering of Ludong University (2008)
文摘This paper considers the pricing problem of collateralized debt obligations tranches under a structural jump-diffusion model, where the asset value of each reference entity is generated by a geometric Brownian motion and jump with an asymmetric double exponential distribution. Conditioned on the common factor of individual entity, this paper gets the conditional distribution, and further obtains the loss distribution of the whole reference portfolio. Based on the semi-analytic approach, the fair spreads of collateralized debt obligations tranches, i.e., the prices of collateralized debt obligations tranches, are derived.
文摘分析了大样本同质投资组合(LHP)近似方法下基于t-NIG单因子Copula的CDO定价模型,此模型是CDO标准定价模型(单因子Gaussian Copula)的扩展延伸.对模型的扩展是为了更好地解决"相关性微笑"问题,并使定价和市场报价更好的吻合.模型中假设市场共同因子和个体异质因子分别服从t分布和NIG分布,构建t-NIG单因子Copula模型.基于模型在大样本同质情况下计算资产池损失分布,并对Dow Jones iTraxx Europe指数的各分券层定价.另外,分析了数值结果,对不同模型适应CDO各分券层的市场报价的能力进行比较.结果表明:所有的模型都比标准模型更好地适应市场报价,且t-NIG单因子Copula模型定价效果最为理想.
文摘This paper aims to reveal the mechanism of Collateralized Debt Obligations (CDOs) and how CDOs extend the current global financial crisis. We first introduce the concept of CDOs and give a brief account of the de-velopment of CDOs. We then explicate the mechanism of CDOs within a concrete example with mortgage deals and we outline the evolution of the current financial crisis. Based on our overview of pricing CDOs in various existing random models, we propose an idea of modeling the random phenomenon with the feature of heavy tail dependence for possible implements towards a new random modeling for CDOs.
文摘Copula method has been widely applied to model the correlation among underlying assets in financial market. In this paper, we propose to use the multivariate Frechet copula family presented in J. P. Yang et al. [Insurance Math. Econom., 2009, 45：139 147] to price multivariate financial instruments whose payoffs depend on the k^th realization of the underlying assets and collateralized debt obligation （CDO）. The advantage of the multivariate Frechet copula is discussed. Empirical study shows that such copula family gives a better fitting to CDO＇s market price than Gaussian copula for some derivatives.
