R help - Sep 2012 - RV: problems for making grids from lmer models

Dear colleagues:
 
I am working with binary mixed models using lme4 package and everything works
fine until I try to spatialize the fitted models using grids predictors. I have
used the "predict" function from RASTER package and the RSAGA
functionalities, but always I have the same error: "Error in object$xlevels
: $ operator not defined for this S4 class".
 
I think the problem comes from the slot (@) used in lme4 for values derived from
fitted models. In other mixed models (gausssian) from nlme package, the slots
are $ and the "predict" function works fine. How do you think can I
solve this problem?' Is there any other function or package to make grid
maps from fitted glmer models?
 
These are the command lines and error message:
 
fitted_glmer_geo<-predict(grid_stack, fitted_glmer,
type="response")
Error in object$xlevels : $ operator not defined for this S4 class

Thanks in advance, and my best regards,
 
Olga
 
Olga Viedma Sillero
Profesora Ordenacion del Territorio
Departamento Ciencias Ambientales
Facultad Ciencias del Medio Ambiente
Avd/ Carlos III, s/n 45071 Toledo (Spain)
email: olga.viedma@uclm.es
Tel: 925-268800 (ext. 5469)
Tel movil: 675 880 297
Fax: 925 268844

________________________________

De: MARIA OLGA VIEDMA SILLERO
Enviado el: mié 26/09/2012 13:09
Para: bates@stat.wisc.edu; bbolker@gmail.com; maechler@R-project.org
Asunto: problems making grids from lmer models




	[[alternative HTML version deleted]]