[R-sig-ME] refit, update and the like

[Douglas Bates]

On Tue, May 12, 2009 at 4:22 PM, Juan Pedro Steibel <steibelj at msu.edu> wrote:
> Hello everyone,
> I have a questions on refitting models using lmer. Some time ago, I asked
> about refitting the same model (X,Z,R) to many responses and was pointed at
> the refit function as a more efficient alternative to just running lmer()
> function iteratively.
> My question now is, suppose I want to change only one column from X (say
> replace covariate X1 with covariate X2), is there a faster way that just
> re-fitting the model with lmer. update() does not seem to be working for
> doing that in my case. I can update to drop only one random effect from the
> model but when I try to drop other terms, I get an error message.

Could you provide a reproducible example please?