Skip to content
Prev 13556 / 20628 Next

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

svm
I appreciate you running the models in Bayesian packages. I've thought
about going that route, though I have no experience fitting these models in
Stan or JAGS or some other package.

As for the slow compile: I took some advice from the vignette on Cran about
performance optimization (even at the expense of suboptimal estimates). I
wish I knew how to parallelize the model, though that's a different topic.

On Wed, Jul 15, 2015 at 12:52 PM, Paul Buerkner <paul.buerkner at gmail.com>
wrote: