[R-sig-ME] complete separation problem lmer

[Steve Walker]

Vincent Dorie's blme package is an extension of lme4, with Bayesian 
point estimation.  It should be able to help with complete separation, 
if you are willing to use weakly-informative prior distributions.

Steve.

On 2013-03-25 12:43 PM, Giovanni Mancuso wrote:
> Hello Everybody,
>
> I'm dealing with the complete separation problem in a probit regression model.
> As I understand there's an R package (brglm) that implements a bias-reduced estimate for GLMs.
>
> Is there anything similar that works with lmer() function?
> I wish to use lmer() function instead of glm(), but I'm not quite sure if it's possible and how to do it.
>
> Many thanks
> Giovanni
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models