Validation of Normal Inverse Gaussian Distribution for Synthetic CDO Pricing

How to determine the default loss distribution of the whole credit portfolio is the most critical part for pricing CDOs. This paper follows Kalemanova et al (2007) and assesses the pricing efficiency of both one-factor Gaussian Copula model the Normal Inverse Gaussian (NIG) Copula model during the turbulent market condition by using data in 2008 and 2009. In addition, we test the price impact of the skewed NIG distribution by adjusting the value of the two parameters. The results show that NIG Copula performs much better than Gaussian Copula, and the introduction of the asymmetry factor in NIG distribution can further improve the modeling results.