The first place to look is surely under "Point pattern analysis"
on the Spatial task view (http://cran.csiro.au/web/views/)
Note especially the spatstat package.
But I am puzzled also. A model that takes account os spatial
correlation might I suppose be described as some kind of
'spatial generalized linear mixed model'
Do you have something more in mind? And why the Laplace
approximation in particular?
John Maindonald email: john.maindonald at anu.edu.au
phone : +61 2 (6125)3473 fax : +61 2(6125)5549
Centre for Mathematics & Its Applications, Room 1194,
John Dedman Mathematical Sciences Building (Building 27)
Australian National University, Canberra ACT 0200.
http://www.maths.anu.edu.au/~johnm
On 04/03/2012, at 8:28 AM, Douglas Bates wrote:
I am taking the opportunity to cc: this reply to the
R-SIG-Mixed-Models mailing list. Members of that list are often more
knowledgeable and quicker to respond than am I.
On Sat, Mar 3, 2012 at 1:44 PM, <hbaghishani at modares.ac.ir> wrote:
I would like to fit an spatial generalized linear mixed model, for example
to model spatial count responses, by Laplace approximation and by using lme4
R package. I'm very enthusiastic if it is possible to implement this fitting
by glmer in lme4 package?
I'm not exactly sure what a spatial generalized linear mixed model is.
Could you or someone else on the list elaborate, please?