[R-sig-ME] lmer (glmer) stratification of fixed effects and the / (forward slash) command (lme4)

Douglas Bates bates at stat.wisc.edu
Fri Oct 28 23:53:31 CEST 2011


On Wed, Oct 26, 2011 at 1:10 PM, Derek Dunfield <dunfield at mit.edu> wrote:
> Hello everyone,
>
> It looks like the / command in lmer  (or glmer), when used for fixed
> effects, is fitting a stratified model.

Actually this question isn't specific to lmer or glmer.  The
fixed-effects part of the formula is exactly the same as that for
linear models.

I'm not sure where the best place to read about the effect of / in a
linear model formula is.  I might try Venables and Ripley's, "Modern
Applied Statistics with S" (aka MASS)

> For example, we have two IVs A and B that we are interested in modeling as
> fixed effects. A has two categories A1 and A2. We will also model the random
> effect R.
>
> Model:  Y ~  A/B + (1|R)
>
> This model produces fixed effect estimates for A, A1:B, and A2:B
>
> It seems that here A1:B and A2:B are stratified fits, with B fit for the A1
> sample set and A2 sample set separately.
>
> Is this correct? If not, what is the syntactic sugar "/" doing?
>
> Thanks in advance for your help.
> Derek
> --
> Derek Dunfield, PhD
> Postdoctoral Fellow, MIT Intelligence Initiative
> Sloan School of Management
> MIT Neuroeconomics Research Center, Prelec Lab
> 77 Massachusetts Ave E62-585
> Cambridge MA 02139
>
>        [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>




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