Comments on: Wired on the Gaussian copula https://blog.yhuang.org/?p=164 here. Tue, 14 Oct 2025 11:10:14 +0000 hourly 1 http://wordpress.org/?v=3.1.1 By: Ninfa https://blog.yhuang.org/?p=164&cpage=1#comment-113522 Ninfa Sat, 08 Dec 2012 19:10:12 +0000 http://scripts.mit.edu/~zong/wpress/?p=164#comment-113522 Only wanna input on few general things, The website design and style is perfect, the subject matter is real fantastic : D. Only wanna input on few general things, The website design and style is perfect, the subject matter is real fantastic : D.

]]> By: David Crow https://blog.yhuang.org/?p=164&cpage=1#comment-92146 David Crow Thu, 11 Nov 2010 07:08:16 +0000 http://scripts.mit.edu/~zong/wpress/?p=164#comment-92146 Great post. Li did not invent the Gaussian copula, he merely applied it to an important finance problem. I haven't read the article (yet), but I think the value of Gaussian copulas resides in their ability to model bi- or multi-variate relationships when each of the marginal distributions is Gaussian, but their joint distribution is not. One example of this is "tail dependency", when extreme events influence one another more than a bivariate normal distribution (with the relationship described by the linear correlation parameter) would predict. This is presumably the case with credit defaults. Great post. Li did not invent the Gaussian copula, he merely applied it to an important finance problem.

I haven’t read the article (yet), but I think the value of Gaussian copulas resides in their ability to model bi- or multi-variate relationships when each of the marginal distributions is Gaussian, but their joint distribution is not. One example of this is “tail dependency”, when extreme events influence one another more than a bivariate normal distribution (with the relationship described by the linear correlation parameter) would predict. This is presumably the case with credit defaults.

]]>