Date:         Wed, 28 Jan 1998 08:23:05 +0100
Reply-To:     Hans-Peter Piepho <piepho@WIZ.UNI-KASSEL.DE>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         Hans-Peter Piepho <piepho@WIZ.UNI-KASSEL.DE>
Subject:      Re: Restricting beta weights
Comments: To: mullenr@EXT.MISSOURI.EDU
Content-Type: text/plain; charset="us-ascii"

>     Restricting beta weights
>
>     I am determining the weights to assign a series of job classification
>     variables. The dependent variable is target salary. The problem I have
>     is that regression solutions are resulting in negative beta weights
>     and a negative weight are unacceptable because they are too hard to
>     explain. I tried to use RESTRICT and discovered that > and < signs
>     were not allowed. Is there a way to find the best least squares fit
>     for a model of positive beta weights?
>
>     Please respond to:
>
>     Steve Chatman
>     Director of Analytical Services
>     University of Missouri - System
>     Columbia, Missouri
>     email: chatmans@ext.missouri.edu
>
>
You could try NLIN, which has a BOUNDS statement allowing < and >
restrictions on the parameters.
_______________________________________________________________________
Hans-Peter Piepho
Institut f. Nutzpflanzenkunde  WWW:   http://www.wiz.uni-kassel.de/fts/
Universitaet Kassel            Mail:  piepho@wiz.uni-kassel.de
Steinstrasse 19                Fax:   +49 5542 98 1230
37213 Witzenhausen, Germany    Phone: +49 5542 98 1248