R help - Oct 2013 - changing model matrix restriction in lm()

Dear useRs,
I was wondering if there was a way of changing the model matrix restriction
automatically in the formula statement when fitting a model using, for example,
lm().
When we do

set.seed(100)
y <- rnorm(12)
A <- gl(3, 4)
summary(lm(y ~ A))

we obtain A1=0 as a baseline and A2 and A3 as effects.
However, if I want to use A2=0 as a baseline, I can do

X <- cbind(1, model.matrix(lm(y ~ A - 1)))[,-3]
summary(lm(y ~ X))

I wonder if there is a way of specifying the restriction in the parameters
directly in the formula argument instead of building the desired model matrix.

Best wishes,
Rafael.

?contrasts
?C

Cheers,
Bert

On Wed, Oct 23, 2013 at 6:00 AM, Rafael Moral
<rafa_moral2004 at yahoo.com.br> wrote:
> Dear useRs,
> I was wondering if there was a way of changing the model matrix restriction
automatically in the formula statement when fitting a model using, for example,
lm().
> When we do
>
> set.seed(100)
> y <- rnorm(12)
> A <- gl(3, 4)
> summary(lm(y ~ A))
>
> we obtain A1=0 as a baseline and A2 and A3 as effects.
> However, if I want to use A2=0 as a baseline, I can do
>
> X <- cbind(1, model.matrix(lm(y ~ A - 1)))[,-3]
> summary(lm(y ~ X))
>
> I wonder if there is a way of specifying the restriction in the parameters
directly in the formula argument instead of building the desired model matrix.
>
> Best wishes,
> Rafael.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.


-- 

Bert Gunter
Genentech Nonclinical Biostatistics

(650) 467-7374