I was recently told that it is not allowed to include static covariates
in mixed effect models. There should be something written in "Applied
Longitudinal Analysis, Fitzmaurice, Laird and Ware, Wiley & Sons, 2004.
Chapter 12", however I do not have the book and cannot look it up.
When I have time series data, and season is a grouping factor, then the
the response mean values should be tested, against the static covariates
in the different seasons or not?
If it is not allowed, how could the effect of the static covariates then
be tested for within a Longitudinal framework? only with a repeated
measures anova?
Jens