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

Re: st: Testing equality of coefficients, linear and quadratic terms

From   Nick Winter <>
Subject   Re: st: Testing equality of coefficients, linear and quadratic terms
Date   Wed, 09 Nov 2005 22:21:25 -0500


You want, I think, a joint test that age1=age2 AND agesquare1=agesquared2

The -accum- option of test will let you construct this.

eg, using the auto dataset

gen mpgsq = mpg^2
gen f_mpg = foreign*mpg
gen f_mpgsq = foreign*mpgsq

reg price mpg mpgsq f_mpg f_mpgsq foreign
--output omitted--
test mpg = f_mpg
--output omitted--
test mpgsq = f_mpgsq, accum

 ( 1)  mpg - f_mpg = 0
 ( 2)  mpgsq - f_mpgsq = 0

       F(  2,    68) =    3.02
            Prob > F =    0.0554

--Nick Winter

At 09:15 PM 11/9/2005, you wrote:
I am familiar with the "test" procedure in Stata to
test for equality of coefficients across independent
samples, though have a question about the command. I
would like to test for equality of two coefficients
across samples, one for AGE and one for a square of
the AGE term (call it AGESQUARED). Basically I would
to determine if age curves are similar across samples,
and would appreciate your insights about how to
incorporate the squared term. Which of the following
would be appropriate (if either)?




Thanks in advance for your help,

Yahoo! Mail - PC Magazine Editors' Choice 2005
*   For searches and help try:
Nicholas Winter                           607.255.8819 t
Assistant Professor                       607.255.4530 f
Department of Government     e
308 White Hall   w
Cornell University
Ithaca, NY 14853-4601

*   For searches and help try:

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