Skip to content
Prev 15476 / 20628 Next

wider than expected confidence intervals with lsmeans and predict.glmmadmb

I will look this over more carefully if/when I get time, to see what
the differences are between the ways in which lme4, glmmADMB, and
glmmTMB implement the building blocks that lsmeans uses. But I can't
guarantee I will get to it soon ... essentially, I'll just have to
pick through the implementation of lsm.basis for glmmTMB and glmmADMB
and see what the differences are ... it would help to try this on some
cases without zero-inflation and without any random effects at all, to
see where the discrepancies are coming from.
On Sun, May 28, 2017 at 5:20 PM, Evan Palmer-Young <ecp52 at cornell.edu> wrote: