The one-factor Gaussian copula method has become the de facto standard to analyze most synthetic collateralized debt obligation structures. Unfortunately, this method produces a peculiar phenomenon known as a correlation smile (the implied correlation determined by the model depends on the CDO tranche one is considering instead of being tranche-independent). Market participants are divided regarding this issue. Many suspect that the correlation smile is caused by a flaw in the above-mentioned modeling strategy although they have been unable to articulate why. Others insist that the smile is actually correct and reveals important and relevant tranche-dependent characteristics, but have failed to produce convincing evidence to support this view. In this article the authors present evidence that the correlation smile is really a by-product (artifact) of an unfortunate modeling strategy and has no financial or market-driven interpretation whatsoever. Moreover, the authors argue that this modeling approach should be abandoned at once.
Cifuentes, Arturo, and Georgios Katsaros. "The One-Factor Gaussian Copula Applied to CDOs: Just Say NO (or, If You See a Correlation Smile, She Is Laughing at Your 'Results')." Journal of Structured Finance 13, no. 3 (Fall 2007): 60-71.
Each author name for a Columbia Business School faculty member is linked to a faculty research page, which lists additional publications by that faculty member.
Each topic is linked to an index of publications on that topic.