[R] model.matrix for non-hierarchical models

[Lukas Meier]

Peter Dalgaard writes:
 > Lukas Meier <meier at stat.math.ethz.ch> writes:
 > 
 > > Dear R-users,
 > > 
 > > Using the function model.matrix I noticed the following
 > > behaviour (example below): Using the formula "~ a + a:b" will
 > > give a matrix of the same dimension as using "~ a * b". In the
 > > first case there are additional columns for the interaction (to
 > > compensate for the missing main-effect).
 > > 
 > > dd <- data.frame(a = gl(3,4), b = gl(4,1,12))
 > > model.matrix(~ a*b, dd, contrasts = list(a="contr.sum", 
 > >              b="contr.sum"))
 > > model.matrix(~ a + a:b, dd, contrasts = list(a="contr.sum", 
 > >              b="contr.sum"))
 > > 
 > > Is there any way to get the design matrix corresponding to "~ a
 > > + a:b" or do I have to do this manually?
 > 
 > The design matrix corresponding to ~ a + a:b is what you go above. If
 > that wasn't what you wanted, perhaps you should tell us what you did
 > want.

The problem is that when using  "~ a + a:b" a model matrix with
the same number of columns like the matrix for the full model
(~a*b) is created. I'd like that when using "~ a + a:b" I get
the same matrix as the full model up to the deletion of the
columns corresponding to the main-effect of b. 

Regards,

Lukas Meier