Re: contrained linear regression

[Prof Brian Ripley <ripley@stats.ox.ac.uk>]

?nnls.fit  may well help

On Wed, 27 Aug 2003 Nicola.Beghetto@bancaimi.it wrote:

> Hi all,
> I need to perform a linear regression in which I have to include the
> constraint that all the final estimated parameters have to be positive.
> I seem not to have seen anything about it on the help files, and I wonder
> if someone of you have written a code for it

-- 
Brian D. Ripley,                  ripley@stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595