Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: re: constraints

From   "Vidya Mahambare" <>
To   <>
Subject   Re: st: re: constraints
Date   Wed, 29 Jun 2005 07:39:46 +0100

Following Kit's suggestion I redefined the constraint   L.wpi+L2.wpi+L3.wpi=1 as
b(L3) = 1-b(L)-b(L2), and regressed 
(wpi - L.wpi)  on (L.wpi-L3.wpi), (L2.wpi-L3.wpi) and other variables.

This does not produce a correct R^2, as Richard suspected. It is much lower
because of variables transformations, I think. Is there a way to to calculate a
correct R^2 after using -reg- for this problem?


>>> 06/29/05 5:20 AM >>>
At 06:45 PM 6/28/2005 -0400, Kit Baum wrote:
>Although cnsreg is useful, it is hardly necessary for Vidya's problem:
>"I am trying to estimate an equation similar to this -
>constraint define 1 L.wpi+L2.wpi+L3.wpi=1
>cnsreg wpi L.wpi L2.wpi L3.wpi L.ygapm L2.ygapm sswpi L.sswpi, constraint(1)"
>This constraint may be rewritten as b(L3) = 1-b(L)-b(L2)  and merely 
>substituted into the equation,
>(wpi - L.wpi)   regressed on (L.wpi-L3.wpi),  (L2.wpi-L3.wpi) and the rest
>which you can estimate with regress, robust.
>If you want the coefficient on L3.wpi, calculate it via lincom.

Since I made the claim that many/most/all problems involving (linear) 
constraints could be handled by regular regression as well as by cnsreg, I 
am glad Kit was able to do the algebra on this one!  However, does this 
produce a correct R^2, which is what Vidya was seeking in the first 
place?  I was thinking of situations where all the manipulations could be 
done on the RHS, but here the LHS has to be changed.

Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):
WWW (department):

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index