[R-sig-ME] lmer and warming message when using method=ML

[Douglas Bates]

On Mon, Nov 24, 2008 at 8:20 AM, Renwick, A. R. <a.renwick at abdn.ac.uk> wrote:
> When you use a lmer model you have to use the command REML=FALSE instead of method=ML.

> One thing to note also is that when you use an lmer model in at not possible at present to use the REML method.

I think you mean that REML is not available for a glmer model (i.e. a
generalized linear mixed model).  For an lmer model REML is the
default and, as you indicated, you get ML estimates by stating REML =
FALSE in your call to lmer.

> Doug Bates replyed to my question asking just that on the 17th Sep "The REML criterion is not defined for generalized linear mixed models. Well, at least I don't know of a definition that makes sense for GLMMs. "

> Hope that helps,
> Anna
>
> -----Original Message-----
> From: r-sig-mixed-models-bounces at r-project.org [mailto:r-sig-mixed-models-bounces at r-project.org] On Behalf Of Maria Jose Juan Jorda
> Sent: 24 November 2008 13:53
> To: r-sig-mixed-models at r-project.org
> Subject: [R-sig-ME] lmer and warming message when using method=ML
>
> Hello lmer experts
>
> In the past the lmer function allowed to choose either ML or REML as the estimation method.
>
> I just updated the R version  to *R version 2.8.0* (2008-10-20)
> (i386-pc-mingw32)  and the lme4 package to *lme4_0.999375-27.
>
>
> *I am running my old code choosing ML as the estimation method and I am getting this warming:
>
> Warning message:
> In lmer(meansize ~ year + (year | id), data = cas, method = "ML") :
>  Argument 'method' is deprecated.  Use 'REML' instead
>
>
> This is my code:
>
> model.a<-lmer(meansize~year+ (year|id),data=cas, method="ML")
>
>
> I have tried to look for the asnwer myself in the R-sig-mixed message board and I could not find an answer.
>
> Does anybody understand the reason why ML cannot be used anymore?
>
> Thanks in advance.
>
> Maria
>
>        [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>
>
> The University of Aberdeen is a charity registered in Scotland, No SC013683.
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>