Going into exploratory data analysis with the intention of fitting copula models, I was looking at the famous copula and they mention here that either contour or 3D plots of both variables, against some third metric, are often used to determine the correct copula to act as a pair for two variables.

Here are some I have generated from the following copulas: two normal dist.s with 0 and 0.5 correlations, respectively, and two t-dist.s with the same respective correlations, with 10 degrees of freedom.

However, looking at these, how precisely does one interpret them? I have only seen this visual approach taken in this and another copula package, VineCopula, and I would like to learn a good resource so I can further read up on this.

Thank you ahead of time for any advice.

Your Answer

 
discard

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.