[R] Need expert help with model.matrix

[Gabor Grothendieck]

On Wed, May 18, 2011 at 7:15 AM, Axel Urbiz <axel.urbiz at gmail.com> wrote:
> Dear experts:
>
> Is it possible to create a new function based
> on stats:::model.matrix.default so that an alternative factor coding is used
> when the function is called instead of the default factor coding?
>
> Basically, I'd like to reproduce the results in 'mat' below, without having
> to explicitly specify my desired factor coding (identity matrices) in the
> 'contrasts.arg'.
>
> dd <- data.frame(a = gl(3,4), b = gl(4,1,12))
> ca <- contrasts(dd$a, contrasts= FALSE)  # 3 x 3 identity matrix
> cb <- contrasts(dd$b, contrasts= FALSE)  # 4 x 4 identity matrix
> mat <- model.matrix(~ a + b, dd, contrasts.arg = list(a=ca, b=cb))
>
> My approach was to modify the code in model.matrix by explicitly setting the
> contrasts argument in the contr.identity and contrasts function to FALSE.
> This is shown at the bottom of the email in the function model.matrix2:
>
> contr.identity <- contr.treatment
> formals(contr.identity)$contrasts <- FALSE
>
> contrasts <- contrasts
> formals(contrasts)$contrasts <- FALSE
>
> However, I believe this function is using contrasts = TRUE, as it doesn't
> return the identity contrasts
> mat2 <- model.matrix2(~ a + b, dd)
>
> Any help here is much appreciated.
> Axel.
>

If your objective in all this is ultimately to get lm coefficients in
the original coding then see ?dummy.coef

-- 
Statistics & Software Consulting
GKX Group, GKX Associates Inc.
tel: 1-877-GKX-GROUP
email: ggrothendieck at gmail.com