Hi @bjoernholzhauer and @weberse2 , thanks a lot for the nice MMRM chapter!

I would have a few improvement ideas about details of the mmrm package description, including the idea about the package's nature not being an lme4 wrapper (here) and the availability (meanwhile) of arm/group specific covariance estimation here.

Would it be ok to propose this in a PR? Otherwise I could also provide more details here.

Thanks,
cheers
Daniel