maechler at stat.math.ethz.ch
Wed May 14 15:39:10 CEST 2003
>>>>> "Giovanni" == Giovanni Marchetti <gmm at ds.unifi.it>
>>>>> on Sat, 3 May 2003 23:48:48 +0000 writes:
Giovanni> I recently posted a question concerning an
Giovanni> inconsistency of expand.grid in defining the
Giovanni> reference level of the factors.
> > expand.grid(x = c("b", "a"), y = c(1, 2))$x
>  b a b a
> Levels: b a # reference level is b
> > expand.grid(x = c("b", "a"))$x
>  b a
> Levels: a b # reference level is a
Giovanni> Thank you very much for the ready explanations and
Giovanni> I found this inconsistency working with
Giovanni> contingency tables and logistic models.
I've forgotten to tell you (and the audience) that this has been
chnaged in R-patched (to be 1.7.1) where
expand.grid(x1) and expand.grid(x1,x2) will be consistent.
Thank you for your posting!
More information about the R-help