[R-sig-ME] Working formula for glmm model in R (bernoulli response)

[Phillip Alday]

Most packages in R lump Bernoulli and Binomial families together, so
this should work:

library(lme4)

glmer(occuppied ~ 1 + year + (1|territory), data=yourdata, family=binomial)


On 19/3/20 2:14 pm, Michael Romanov via R-sig-mixed-models wrote:
> Dear all,
>  
> Sorry for the silly question, but I got stuck on it.
>  
> My dataset is occupancy of eagle territories in different years (see the example below). I’m trying to test my data for time trend, taking into account territory quality (as a random effect).
>  
> territory year occupied
> CHV-1 2006 0
> CHV-12 2006 0
> CHV-120 2006 1
> CHV-13 2009 0
> CHV-14 2009 1
> CHV-15 2010 1
> CHV-16 2010 1
>  
> My thoughts are following. ‘year’ is a fixed effect variable and ‘territory’ is a random effect variable. For example, in lme4 package the formula could be following: occupied = year + (1|territory). However, lme4 doesn’t support Bernoulli family, so I chose glmm package. According to its specifications, the glmm function accepts two different formulae, separately for fixed and random effects:
>  
> glmm(fixed, random, varcomps.names, data, family.glmm, m, varcomps.equal, doPQL = TRUE,debug=FALSE, p1=1/3,p2=1/3, p3=1/3, rmax=1000,iterlim=1000, par.init, zeta=5, cluster=NULL)
>  
> So now I need two different formulae. I tried ‘occupied ~ year’, ‘occupied ~ territory’, but it doesn’t work. Namely, the RStudio gets into an infinite loop and doesn’t produce any output.
>  
> Can anyone help me to write a working formula (or formulae) to properly run the glmm model?
>  
> Thanks,
>  
> Michael
>  
>  
>  
>  
>  
>  
>  
>  
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models