[R-sig-ME] Model building problem?

[David Duffy]

On Sat, 13 Mar 2010, Luciano La Sala wrote:

> Hello everyone, 
> I am building a model using the lmer function. I have IgG 
> (continuous) as my outcome of interest, 
> and the following variables as fixed effects: Egg Breadth (continuous), i
> Egg Length (continuous), EggVolume (continuous), 
> Clutch Size (three levels), and Hatching Order (three levels), plus random intercepts for NestID.

logLik Model
282.1  ELISA2~EggBreadth+EggLength+ClutchSize+HatchOrder+ EggVolume+(1|NestID) 
286.2  ELISA2~EggBreadth+EggLength+HatchOrder+EggVolume+(1|NestID) 
289.1  ELISA2 ~ EggBreadth + HatchOrder + EggVolume + (1 | NestID) 
287.6  ELISA2 ~ EggLength + HatchOrder + EggVolume + (1 | NestID) 
293.7  ELISA2 ~ EggBreadth + EggVolume + (1 | NestID) 
297.2  ELISA2 ~ EggBreadth + (1 | NestID) 
298.8  ELISA2 ~ 1 + (1|NestID)

I am feeling pretty dull today, but I don't expect the model likelihood to
increase as the model is simplified (I always get confused when the sign 
on the log likelihood is positive). IgG is usually log-normal, has it been
transformed appropriately?

-- 
| David Duffy (MBBS PhD)                                         ,-_|\
| email: davidD at qimr.edu.au  ph: INT+61+7+3362-0217 fax: -0101  /     *
| Epidemiology Unit, Queensland Institute of Medical Research   \_,-._/
| 300 Herston Rd, Brisbane, Queensland 4029, Australia  GPG 4D0B994A v