iitis / datagencopulabased.jl Goto Github PK

Hey, kind of new to the whole copula thing. Could you show an example of how to get the empirical copula for a given dataset?

Any help or advice appreciated. Love how thorough the docs are.

Hello! I am a user/developer trying out this package for use in EconomicScenarioGenerators.jl and have some questions/feedback. Thank you for this package ❤️

I think that GaussianCopula and MarshalOlkinCopula would be more Julian names than Gaussian_cop and Marshal_Olkin_cop, for example.

https://docs.julialang.org/en/v1/manual/style-guide/#Use-naming-conventions-consistent-with-Julia-base/

Hey,

Great package. i'd like to simulate some clayton copula on BigFloat precision, but your methods restrict to Float64.

Would you consider switching the restriction to Real ? I dont think anything else should mbe changed, just wave the restriction and it'll work for free :)

Hi, very nice library.
I am noticing that there is no way to actually pass to the copula generator a prescribed random number generator (like in randn, rand etc).
It would be very nice to have such a functionality.

Hi again,

Following the examples from the base Random library, it would be nice to have a simulate_copula! function where the first argument is the output Matrix instead of the integer representing the number of simulations. This will lead to a great performance improvement (IMHO).

This issue is used to trigger TagBot; feel free to unsubscribe.

If you haven't already, you should update your TagBot.yml to include issue comment triggers.
Please see this post on Discourse for instructions and more details.

If you'd like for me to do this for you, comment TagBot fix on this issue.
I'll open a PR within a few hours, please be patient!

As a suggestion, not really an issue, I would suggest to use "empty" types for dispatching the correlation type instead of Strings, in order to clean up the implementation.

A PkgEval run for a Julia pull request which changes the generated numbers for rand(a:b) indicates that the tests of this package might fail in Julia 1.5 (and on Julia current master branch).

Also, you might be interested in using the new StableRNGs.jl registered package, which provides guaranteed stable streams of random numbers across Julia releases.

Apologies if this is a false positive. Cf. https://github.com/JuliaCI/NanosoldierReports/blob/ab6676206b210325500b4f4619fa711f2d7429d2/pkgeval/by_hash/52c2272_vs_47c55db/logs/DatagenCopulaBased/1.5.0-DEV-87d2a04de3.log

In order to use your library from outside it would be very nice to have kind of a dispatch functionality.
Currently in order to specify which copula to use you need to call different functions, this is not so useful when you want to write generic code on top of your library (my case XD).

It would be nice to select the kind of copula though a struct/type which carries as well the copula parameters, and then a generic simulate_copula method which will dispatch to the correct copula.

I was going to implement this from outside but probably it's better if done already in the library.

Just to give you a proper example:
Current:
Corr=[1.0 0.0;0.0 1.0];
U_joint=gausscopulagen(Nsim,Corr);
Proposal:
copula_=GaussianCopula([1.0 0.0;0.0 1.0]);
U_joint=simulate_copula(Nsim,copula_);