Copula

[ravishankar80]

Hi,

I read a interesting article about Copulas and it's role in the present crisis.Pasting the link below.Should be of interest to everyone

http://www.wired.com/techbiz/it/magazine/17-03/wp_quant?currentPage=all



Regards
Ravi

 

[David Harper CFA FRM]

Thanks Ravi,

the comments are just as interesting as the article! It intersts me that some models and metrics (VaR, copula) have become popular targets; and you get the sense, in some cases, they are blamed merely because are deemed too complex. I really like what the 2nd commentor says, "First off, there is nothing wrong with a model. Models are good. They offer a mode of understanding the world where a law is lacking. For example, we have hundreds of competing climate change models that explain what is going on currently, and predict what we should expect. But, you must know the limitations of your model. By definition, a model is a simplification of a complex phenomenon. That does not make it flawed: that makes it a model."

Because i agree with that comment, I personally find it impossible to attach any value judgement to any model. The issue is about application, and it seems to me it is impossible to make assessments without a threshold understanding of the model. In my feedback to GARP, I requested they add copulas formally to the agenda. Like so many tools, there is a positive/negative learning. Positive: here is a model, here is how is works. Negative: here are its special issues in practice. Thanks for sharing, David