Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Testing the extent of difference between two coefficients in the same model
From
Richard Williams <[email protected]>
To
[email protected], [email protected]
Subject
Re: st: Testing the extent of difference between two coefficients in the same model
Date
Tue, 31 Jan 2012 11:19:22 -0500
At 06:34 AM 1/31/2012, Maarten Buis wrote:
On Mon, Jan 30, 2012 at 11:12 PM, David Hoaglin wrote:
> A model in
> which the two coefficients are equal is the same as the model in which
> the predictor associated with that coefficient is the sum of "old" and
> "new."
This is a trick that is often forgotten (at least in my discipline),
but can have surprisingly useful applications beyond adding linear
constraints to models that do not allow the -constraint- option. I
submitted a Stata tip on this topic a couple of days ago.
<http://www.maartenbuis.nl/wp/sumconstr.html>
-- Maarten
As Maarten's tip shows, this approach lets you do likelihood ratio
tests of the constraint, whereas the -test- command does a Wald test.
Starting with Stata 11, the -logit- and -ologit- commands allowed the
use of the -constraint- option, so the tip isn't quite as necessary
as it used to be, although I think it may be just as easy if not
easier to compute the sum as it is to specify the constraints. For
teaching purposes I like to write out the equations and illustrate
approaches like this just so my students understand the constraints
being imposed and why this works.
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME: (574)289-5227
EMAIL: [email protected]
WWW: http://www.nd.edu/~rwilliam
*
* 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/