[R-sig-ME] lme4a, glmer and all that

[Douglas Bates]

On Thu, Mar 4, 2010 at 11:32 AM, Federico Calboli
<f.calboli at imperial.ac.uk> wrote:
> On 4 Mar 2010, at 16:11, Douglas Bates wrote:
> <cut>
>>
>> which, now I see are more consistent with the results from lme4, not
>> lme4a.  I misunderstood the message I received regarding the Stata
>> results.
>>
>> OK, this might have all been a red herring.  I'll look into it some more.
>>
>> At one time on Saturday Night Live Gilda Radner would play a citizen
>> commentator on weekend news update who was confused about the issue
>> and, after a long harangue, would end up realizing she got it wrong.
>> She ended by saying "Never mind".  That might be the case here too.
>
>
> Just out of curiosity, how do we know stata and lme4 are correct and lme4a is not? why is stata the benchmark? Please note I am only being facetious.

In cases like this we usually go with majority rule.  I'm still
verifying and validating.

> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>