[R-sig-ME] lmer: problem in crossed random effect model with verydifferent variances
bates at stat.wisc.edu
Wed Jun 17 22:04:14 CEST 2009
On Wed, Jun 17, 2009 at 2:56 PM, Luca Borger<lborger at uoguelph.ca> wrote:
>> I probably should have used log-transformed y.
> does the discrepancy between lmer&SAS persist if using log(y) (and are the
> distributional assumptions of the model reasonably met with the
> log-transformed response?). Furthermore, I think SAS and lme4 use different
> algorithms, which might contribute to differences in the estimates.
I imagine they do but, because I don't know what SAS does, I can't
say. As soon as SAS Institute goes Open Source we will be able to
make a meaningful comparison :-)
> ----- Original Message ----- From: "Michael Li" <wuolong at gmail.com>
> To: <r-sig-mixed-models at r-project.org>
> Sent: Wednesday, June 17, 2009 3:15 PM
> Subject: [R-sig-ME] lmer: problem in crossed random effect model with
> verydifferent variances
>> Hi, I remember seeing this mentioned somewhere but couldn't find it.
>> I used lmer to fit a simple linear mixed model with two crossed random
>> effects, day and analyst, with no other fixed effects. So the syntax
>> is something like:
>> lmer (y ~ (1 | day) + (1 | analyst), data = data)
>> I can also fit the same model in PROC MIXED. Most of the time I got
>> the same answers. But there seems to be a problem with lmer when one
>> of the random effect has a much smaller variance than others.
>> In my case, SAS would give random effect variances of 1552, 599133 and
>> 213814 for analyst, day and residual effects, respectively but lmer
>> gives 2x10^-12, 599050, and 214680. Basically all parameter estimates
>> are the same (more or less), except that lmer gives very tiny estimate
>> for the random effect of 'analyst'.
>> I probably should have used log-transformed y. But aside from that,
>> how can I get lmer to give a sensible answer? Or is SAS giving the
>> right answer?
>> R-sig-mixed-models at r-project.org mailing list
> R-sig-mixed-models at r-project.org mailing list
More information about the R-sig-mixed-models