Emily Butler

sorry - it didn't copy the error message. Here it is Error in terms.default(model) : no terms component nor attribute In addition: Warning message: Variable transformations in the model formula...

Thanks! I really appreciate the great package :) Cheers, Emily On Sun, Dec 9, 2018 at 3:50 PM Jacob Long wrote: > Hi @sag4040 , you're not the first one...

They are all used for both residuals and group effects (I agree with you that is a better term than random - I just have the bad habit due to...

you are correct Jonah, except that auto-correlation doesn't make sense for the group-level effects (at least not in any application I've seen or thought of).... And just for some context,...

thanks! Yes, that would be great. And I just thought of one more, which is usually called "variance components" and has variances for each effect, but no covariances. I think...

thanks - that's what I thought.... :) On Sun, Apr 15, 2018 at 10:47 AM, Paul-Christian Bürkner < [email protected]> wrote: > Yes indeed. Instead of | use || in group-level...

so do these recent thoughts suggest you are working on implementing correlation structures???? I am close to needing to decide what I will cover in my class in the fall,...

Sounds hopeful :) I am mostly interested in them for the residuals, because then it would give most of the functionality of *gls*, which is a real work-horse in my...

hi again - Same time next year and I am again working on updating my course materials :) Any chance any further work has been done on correlation structures for...

thanks for thinking about it! Compound symmetric and unstructured residuals would be the highest on the wish list. And many thanks again and again for your awesome contribution to the...