Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Maarten Buis <maartenlbuis@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: cnsreg-Constrained Weighted OLS |
Date | Wed, 21 Dec 2011 20:53:48 +0100 |
On Wed, Dec 21, 2011 at 7:46 PM, Martin wrote: > I have following problem. I try to run a regresion like this: > > y=b0 + b1*w1*x1+b2*w2*x2 +b3*w3*x3+b4*w4*x4 > > b0=intercept, b1,b2,b3,b4 Regressors w(i) Weights of the Variable X(i). In > my problem these w are market values of a country and x are Dummies (if > y=stock return is of this country). So because of multicolinearity I have to > put 2 constraints: > > (1) w1*b1+w2*b2=0 > (2) w3*b3+w4*b4=0 Are the weights observed or do you want to estimate them? If they are abserved, just create new variables that are the products of the xs and their weights. In that case no constraints are necessary. If they are to be estimated, see <http://www.maartenbuis.nl/wp/prop.html>. Hope this helps, Maarten -------------------------- Maarten L. Buis Institut fuer Soziologie Universitaet Tuebingen Wilhelmstrasse 36 72074 Tuebingen Germany http://www.maartenbuis.nl -------------------------- * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/