[R] Least Squares Restricted Estimator

[Bert Gunter]

Nelson:

1. Please read the posting guide: your question is practically
unintelligible. You are far more likely to get a useful response if you
follow its recommendations.

2. However, I'll make a stab. **If** what you mean is simply to be able to
choose arbitrary contrasts to estimate in linear models see ?lm, ?contrasts,
?C -- it can already been done. You may also be interested in pp 146-149
(the section on Contrast Matrices)in V&R's MASS Fourth Edition (a most
useful book to have on your shelf, IMHO), which I believe may be exactly
what you're looking for. The gregmisc package/bundle, among others, may also
contain some useful functions for you to fit arbitrary contrasts, I think. 


Cheers,
Bert Gunter
Genentech

-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On
Behalf Of Villoria, Nelson B
Sent: Monday, May 19, 2008 8:25 AM
To: r-help at r-project.org
Subject: [R] Least Squares Restricted Estimator

It is there any implementation of a Least Squares Restricted Estimator
for a single equation in R? I have seen in the list some examples in
which linear constraints are embedded within the equation, but let's say
that I have a large number of coefficients (on factors) that I want them
to sum up to 0 (so I can include an intercept, the full set of factors,
and avoid the dummy trap). Is there any way to impose the restrictions
in some matrix fashion (i.e. Rb = q)?

 

Thanks,

 

Nelson Villoria

 


	[[alternative HTML version deleted]]

______________________________________________
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.