[R-sig-ME] ML or REML for LR tests
ken at kjbeath.com.au
Tue Sep 23 00:59:39 CEST 2008
On 11/09/2008, at 5:01 AM, Austin Frank wrote:
> First off, thanks to all who have responded to the series of
> questions I
> On Fri, Aug 29 2008, Ken Beath wrote:
>> On 29/08/2008, at 2:47 PM, Austin Frank wrote:
>>> 3) Is it the case that LR tests between REML models with different
>>> random effects are meaningful? Does this apply to both nested and
>>> non-nested models?
>> Maybe, but only for nested (see Q2). Supposedly it works better than
>> ML. The significance tests wont be correct but if there is a huge
>> significance level then there is probably a random effect. Simulation
>> seems a better idea.
> Ken was the only one to address this particular point, and I want to
> make sure I've got it straight. Are REML-based likelihood-ratio tests
> (presumably not performed with anova.mer, as that sets REML=FALSE on
> call to logLik) an acceptable method for testing nested models with
> different random effects specifications?
This is discussed in Pinheiro and Bates. It is not statistically
correct, so probably isn't a good idea for a publication. I recommend
reading the sections in Verbeke and Molenberghs book on Linear Mixed
Models. I have used AIC.
The easiest way to avoid choices is to decide that certain parameters
must be modelled by random effects based on medical, biological etc
> As a point of reference, the anova() method is called on two lmer
> that differ only in their random effects in the manuscript by Baayen,
> Davidson, and Bates at
> 12-15). The discussion of that analysis makes no mention of the
> difference between REML and ML fits. Is this because, as discussed
> recently, the REML and ML estimates are so close that there is no
> practical difference in which quantity is used for this test?
> Thanks again!
> Austin Frank
> GPG Public Key (D7398C2F): http://aufrank.net/personal.asc
> R-sig-mixed-models at r-project.org mailing list
More information about the R-sig-mixed-models