[Rd] Different behavior of model.matrix between R 3.2 and R3.1.1

[Frank Harrell]

Terry Therneau has been very helpful on r-help but we can't figure out 
what change in R in the past months made extra columns appear in 
model.matrix when the terms object is subsetted to remove stratification 
factors in a Cox model.  Terry has changed his logic in the survival 
package to avoid this issue but he requires generating a larger design 
matrix then dropping columns.

A simple example is below.


strat <- function(x) x
d <- expand.grid(a=c('a1','a2'), b=c('b1','b2'))
d$y <- c(1,3,2,4)
f <- y ~ a * strat(b)
m <- model.frame(f, data=d)
Terms <- drop.terms(terms(f, data=d), 2)
model.matrix(Terms, m)

   (Intercept) aa2 aa1:strat(b)b2 aa2:strat(b)b2
1           1   0              0              0
2           1   1              0              0
3           1   0              1              0
4           1   1              0              1
. . .

The column corresponding to a='a1' b='b2' should not be there
(aa1:strat(b)b2).

This does seem to be a change in R.  Any help appreciated.


Terms attributes factor and term.labels are:

attr(,"factors")
          a a:strat(b)
y        0          0
a        1          2
strat(b) 0          1
attr(,"term.labels")
[1] "a"          "a:strat(b)"


Frank