Skip to content
Prev 13554 / 20628 Next

Perfectly correlated random effects (when they shouldn't be)

svm
Appreciate all the input. My preference would be for the minimal models for
concern of computation and overparameterization, but my field is one where
the reviewer is always right and junk models are encouraged. I do
appreciate the insight about the small number of regions. I think I had
been mistaken in my belief that the number of countries per region would
compensate even if the number of regions themselves were small.

On Wed, Jul 15, 2015 at 7:32 AM, Reinhold Kliegl <reinhold.kliegl at gmail.com>
wrote: