Paul-Christian Bürkner

Thank you! This look much better. What I don't understand still, is whether this **mathematically** implies the same model? I was under the impression that it does but your last...

Thank you for your time and effort! Perhaps, there is a mistake in the current brms implementation or we messed up the indexing during transformation form z to r in...

Just a note that `brms:::elapsed_time` now also works with cmdstanr and defaults to `NULL` if no information is provided (not `NA`). I think `NULL` would also be a better default...

@fweber144 Do you know if this PR is still relevant for something that will be implemented in projpred in the future?

Thanks for opening this issue. The bug should already be fixed in the github version of brms I think. With regard to the feature request, I agree that it would...

Thanks. I think `by` should be either part of `group` of `coef` or something entirely new. Making it part of `class` would be inconsistent. I will need to think more...

this has not been implemented yet because it needs a new argument in the prior specification functions and I have to think carefully about a name and how that can...

I think that this should be possible. In fact I have thought about this feature for some time already. Given the complexity of such a feature (at least when combined...

Feel free to talk about this now. Maybe it makes sense to implement composites along with CFAs at the same time.

Thanks for the explanation! I think implementing this will require very similar formula parsing than CFAs, so it will likely make sense implementing CFAs and composites together.