Skip to content
Prev 15492 / 20628 Next

Deviations from normality in MCMC models

Is it possible you could use the "exponential" family in mcmcglmm? It seems
to have a (negative) log link function, which could address the skew in
your data.
Reference this thread on exponential family in mcmcglmm
<https://stat.ethz.ch/pipermail/r-sig-mixed-models/2013q3/020851.html>
On Thu, Jun 1, 2017 at 9:47 PM, Ben Bolker <bbolker at gmail.com> wrote: