-
Notifications
You must be signed in to change notification settings - Fork 10
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[New feature] Conditionning #150
Comments
BivariateCopulas.jl
we perform conditioning of copulas and joint distributions using numerical finite differencing. Not sure if there's a better way to do it, but it allows you to sample from arbitrary (without a family) copulas and distributions. I think this is needed if you want to do implement vines
I am writing some functions to implement conditional samplers for copulas where there is a simple analytic formula (e.g FGM copula) as part of another project. Would this be useful to add as a start? I am using the formulae in Appendix 1 of the Trivedi and Zimmerman book. |
@kmarray98 Yes it might ! Even if it does not implement it for all copulas, if you can figure out an interface and API that would provide this functionality, that is great. Do not hesitate to write your code targetting a PR to copulas.jl, I'll help you merging it. |
@kmarray98 I will reconsider this myself, do you have some kind of base on which I could build up ? Even a draft would help |
Hi - sure, sorry I have been quite busy since posting, but I do have something. How is best to transfer it (first time trying to contribute something to a package)? |
Since we finally patched the GlobalSensitivity.jl issues, the urgency is not there anymore. If you want to take more time to pause and ponder before sharing, then it's ok for me too. What's the best sharing way depends on what you have. The most practical is to open a PR: fork the package, push to your own branch and open a PR. If you are not accustomed to Julia package developpement and GitHub Pull Request functionalities however, it can be a bit hard to grasp the first time but it's a tool worth learning ! |
AnderGray said in the review:
So yes conditional density functions for bivariate copulas will be needed to implement vines. I am not sure for the moment how it should be done, and I hope I can do it generically enough, but this is a bit of work.
The text was updated successfully, but these errors were encountered: