Alejandro Catalina

Sorry I took some time to get back to this because of my Christmas holidays! We are pursuing this direction as we speak and already have a prototype to deal...

Yes. This is a bit less straightforward to fix as we can compute both approximate LOO or full LOO, so I think the right weights should be returned by `loo_varsel`...

This happens because we have not implemented parsing support for by syntax in smooth terms. It is desirable to have but we haven’t had the time yet. Best, Alejandro On...

Submodels can be fitted with offsets if I recall correctly. Best, Alex ________________________________ From: Frank Weber ***@***.***> Sent: Monday, August 9, 2021 11:12:57 AM To: stan-dev/projpred ***@***.***> Cc: Subscribed ***@***.***>...

Yes. I see your confusion. We take the offset off the model and keep it in the reference model structure to use it afterwards. In short, we don’t fit the...

I honestly haven’t tried it! I will take a look and see if either it works out of the box or the needed modifications are simple enough! Best, Alejandro On...

Hi @csetraynor and thank you for your interest and work in projpred. We are currently making significant changes to how projpred works (you can take a look at develop branch)...

We are currently studying the latent function approach and it seems very promising for any general model, so it's very likely that we push in that direction. Given that we...

This is a known limitation of the current system and it’s on the todo list, thanks for reporting. From: Frank Weber ***@***.***> Date: Monday, 26. July 2021 at 15.54 To:...

This has to do with the way we split the original fit object's formula to do the forward search. We defined some conventions that seemed reasonable but they may give...