[R-sig-ME] glmer ordered response models

Jarrod Hadfield j.hadfield at ed.ac.uk
Fri Feb 26 21:43:41 CET 2010


Hi,

Just to avoid confusion - "ordinal" is the family argument for ordered  
response models in MCMCglmm not "categorical".  If your response has  
two levels then it makes no difference (except "categorical" uses the  
logit link and "ordinal" uses the probit link). You can fit this in  
glmer using  family = binomial(link="logit") or family =  
binomial(link="probit"). If there are more than two levels I think it  
is not possible to fit the model in glmer, but could be wrong.

Cheers,

Jarrod






Quoting "Doran, Harold" <HDoran at air.org>:

> See ?family for available link functions.
>
> -----Original Message-----
> From: r-sig-mixed-models-bounces at r-project.org   
> [mailto:r-sig-mixed-models-bounces at r-project.org] On Behalf Of jungck
> Sent: Friday, February 26, 2010 2:59 PM
> To: r-sig-mixed-models at r-project.org
> Subject: [R-sig-ME] glmer ordered response models
>
> Hi, all.
>
> I am a new comer to lmer package in R.
> When I run a ordered response model, I couldn't
> figure out what is the option name for "family=" option.
> R keep saying error warnings when I use family="categorical", which
> seem to be used in MCMCglmm package. I don't know how to do bayesian
> model. so I want to know how to do glmer function with ordered dependent
> variable.
>
> Thanks for your help!
>
> -CK
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>
>



-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.




More information about the R-sig-mixed-models mailing list