search for: specificatio

...am amazed! I've been using quite some multilevel/mixed modeling packages, lme4 is a strong candidate for the overall winner, especially for multilevel generzlized linear models. Now go back to my two-level poisson model with cross-classified model. I've been testing various different model specificatios for the past couple of days. Here are the models I tried: 1) Two level random intercept model with level-1 covariates only m1 <- lmer(.D ~ offset(log(.Y)) + (1|provn) +(1|bcohort) + x1 + x2 , data, poisson, method="Laplace") 2) Two-level random intercept model with both level-1 and...

...del simplification. REML will not allow me to remove the fixed factor, so I have to use "regular" ML for this: I think I would prefer to characterize the situation by saying that the log-likelihood from the REML criterion cannot be used to compare models when you change the fixed-effects specificatio. However, this still doesn't explain to me why you want to retain the Treatment factor when you have judged it to be inert. > model4<- lme(DeathDay ~ Treatment, random=~ 1 | Clutch/Cup,method="ML") #Full model > model5<- lme(DeathDay ~ Treatment, random=~ 1 | Clutch,met...