While lm() is a linear modeling, the constraints make it easier to solve with a
nonlinear
tool. Both my packages Rvmmin and Rcgmin (I recommend the R-forge versions as
more
up-to-date) have bounds constraints and "masks" i.e., fixed
parameters.
I am actually looking for example problems of this type that are more recent
than the ones
that got me into this 30 years ago. Do contact me off-list if you have something
that
could be shared. I'd also welcome discussion on appropriate tools for such
constrained
linear modeling problems. They are, I believe, more or less present in most
linear
modeling situations, but we rarely impose the constraints explicitly, and tend
to use lm()
and (hopefully) check if the solution obeys the conditions.
Best,
John Nash
On 06/25/2012 06:00 AM, r-help-request at r-project.org
wrote:> Message: 5
> Date: Sun, 24 Jun 2012 03:34:10 -0700 (PDT)
> From: rgoodman <rosa.goodman at gmail.com>
> To: r-help at r-project.org
> Subject: Re: [R] Constrained coefficients in lm (correction)
> Message-ID: <1340534050627-4634321.post at n4.nabble.com>
> Content-Type: text/plain; charset=us-ascii
>
> Hi Jorge,
>
> Did you ever figure this out? I want to do the same thing with the
> additional constraint of the coef for x1 = 2.
>
> lm(Y~offset(2*x1)+x2+x3,data=mydata)
> where b= coeff for x2, c = coeff for x3, b+c=1 and b and c>0.
>
> I've loaded the systemfit package, but the suggestion "R*beta0 =
q, where R
> is R.restr and q is q.restr in the function call" makes no sense to
me.
>
> Cheers,
> Rosie
