[R-sig-ME] how to know if random factors are significant?
Leonel Arturo Lopez Toledo
llopez at oikos.unam.mx
Wed Apr 2 12:39:09 CEST 2008
Thank you very much all for your comments. Im trying everything you said and checked
previous messages from Marianne, John and Douglas Bates. I found that removing the
random factor reduced the p-values for fixed effects (as John mentioned) and this new
model has not better significant power indicated by the likelihood ration test.
So, even with a low Variance the random factor must be used in a mix model effects.
Thank you very much for your help. I'm understanding and learning so much with the "R-
Este mensaje ha sido analizado por MailScanner
en busca de virus y otros contenidos peligrosos,
y se considera que está limpio.
For all your IT requirements visit: http://www.transtec.co.uk
More information about the R-sig-mixed-models