[R-sig-ME] Slight differences in fitted coefficients in lme4_1.0-6 compared to lme4_0.999999-2

[Ben Bolker]

On 14-02-07 03:34 PM, Tom Wenseleers wrote:
> Dear all, I noticed that I get very slight differences in my current
> lme4 1.0-6 models compared to the old ones I obtained earlier using
> lme4_0.999999-2. I was just wondering whether it would somehow still
> be possible to reproduce the output of the old lme4_0.999999-2, by
> setting appropriate options of the optimizer to use etc? Or is this
> not possible? I also tried installing the old lme4 version using
> install_url in package devel, but if I try this I get a complaint
> that the old version doesn't work with R.0.2. Any easy way to go back
> to the old version (I need this to be able to fully reproduce
> published results)?
> 

  I think you should be able to install lme4.0 from
http://lme4.r-forge.r-project.org/repos/  to reproduce previous outputs.
 You *might* be able to reproduce previous results by setting
control=lmerControl(optimizer="optimx",optCtrl=list(method="nlminb")),
but I don't think we could guarantee that -- too much of the internal
machinery has changed too radically.

  Ben Bolker



> Cheers, Tom
> 
> 
> 
> [[alternative HTML version deleted]]
> 
> _______________________________________________ 
> R-sig-mixed-models at r-project.org mailing list 
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>