statisticalrethinkingjulia / statsmodelcomparisons.jl Goto Github PK

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!

Rob,

Would you like to link to Stan when this is polished up a bit more?

I'm not sure (I am always confused how licenses interact or are compatible with each other) but the comment in

seems to indicate that at least parts of the package have to use the GPLv3? This seems a bit problematic and is not mentioned in the LICENSE.md. I guess it would be possible to resolve this problem by reimplementing the algorithm without looking at the original code (and probably also not the existing implementation in this package if it was based on the original code), eg., just by studying the original paper.

AIC, BIC, and DIC have all been fully supplanted by WAIC, WBIC, and LOO-IC, and there's no longer any real reason to use them. I would suggest removing them from the package to avoid confusing beginners. Someone seeing them used in the package may be led to believe that using them is a good idea, that they provide information different from that provided by the newer information criteria, or that there must be some reason they're in the package. BIC is especially problematic: The name simultaneously gives new users the false impression that the BIC is actually Bayesian and that WAIC/LOO-IC are not Bayesian.

It might be a good idea to implement WBIC to have a replacement for BIC, since BIC has a different use case from the other information criteria in this package (asymptotically maximizing the probability of selecting the correct model, rather than maximizing the expected predictive accuracy of the model). I would also suggest referring to it as the Watanabe-Schwarz Information Criterion to avoid giving the impression that WBIC is "Especially" Bayesian.

Hi Rob,

Please register when everything is to your specification. Let me know if you need me to make any changes.