Skip to content

Alternatives to lmer

1 message · Iker Vaquero Alba

#
?? Yes, that could be a good idea, I definitely have to learn how to use the function "try()". But it's especially annoying this is not working because I have implemented a few similar models before with other dependent variables and all of them worked fine.? 

?? I have attached the data set I am using, hope that helps.

?? Thank you very much.

?? Iker

--- El dom, 16/1/11, Murray Jorgensen <maj at waikato.ac.nz> escribi?:

De: Murray Jorgensen <maj at waikato.ac.nz>
Asunto: Re: [R-sig-ME] Alternatives to lmer
Para: "Iker Vaquero Alba" <karraspito at yahoo.es>
CC: r-sig-mixed-models at r-project.org
Fecha: domingo, 16 de enero, 2011 21:53

Hi Iker,

you don't give us information about your data set or the number of 
levels of your factors. Although you have removed some of the 
interactions from the model it would seem that you are trying to 
estimate a very large number of parameters in the fixed part of the 
model. I would guess that many would be unidentifiable. Any software 
that you might try could have trouble fitting this model. Why don't you 
start off more slowly with main effects only and then use try() to 
identify interactions that would be worth adding to your model?

Murray Jorgensen
On 17/01/2011 9:15 a.m., Iker Vaquero Alba wrote: