Skip to content

Question regarding glmer-function of R-package lme4

2 messages · Douglas Bates, dave fournier

#
As far as I know the REML criterion is not defined for generalized
linear mixed models.  I have taken the liberty of copying the
R-SIG-Mixed-Models at R-project.org list on this response in case others
who read the list can offer suggestions.

On Tue, Jun 14, 2011 at 11:31 AM, Greb, Friederike
<fgreb at uni-goettingen.de> wrote:
#
Operationally it is a fairly simple process to "integrate out"
any fixed effect parameters via the Laplace approximation in  nonlinear
mixed models using AD Model Builder.  This could be used to
provide a general "REML like" estimation procedure. These ideas
are not new,  See

Title: Integrated likelihood methods for eliminating nuisance parameters
Author(s): Berger JO, Liseo B, Wolpert RL
Source: STATISTICAL SCIENCE   Volume: 14   Issue: 1   Pages: 1-22
Published: FEB 1999

Of course how well a particular approach works in practice could
be investigated via simulation.