Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: -gologit2- and testing the difference between coefficients

From   Richard Williams <>
Subject   Re: st: -gologit2- and testing the difference between coefficients
Date   Thu, 29 Mar 2007 12:42:15 -0500

At 08:18 AM 3/29/2007, Herve STOLOWY wrote:
Dear Statalisters:

I use the -gologit2- command (downloadable from ssc). Here is my syntax:

xi, prefix(): gologit2 format size1 auditor_d standards_d for_list sales_int instit_for lev2 i.gics4, v1 nolabel

The dependent variable, format is coded 0, 1 or 2. I get two equations and, consequnetly, two panels of results.

I would like to know if is possible to test, for a given variable, if the two coefficients in each equation are different from one another? To be clearer, I have the variable sales_int. In panel 1, the coefficient is 1.314. In panel 2, the coefficient for the same variable is 1.913. Both are significant at the 0.10 level. Is there a way to test that 1.314 is statistically different from 1.913?

Best regards

Hervé Stolowy
There are various ways to this; as others have said, you can read more on my website. To do specifically what you ask, you can do an LR test, e.g. do something like

gologit2 y x1 x2 x3, store(m1) lrf
gologit2 y x1 x2 x3, pl(x1) store(m2) lrf
lrtest m1 m2

In the first model you are allowing the effects of x1 to differ across panels. In the 2nd model, The -pl- (parallel lines) option will force the effects of x1 to be the same across panels. You can then do a LR chi-square contrast between the two models.

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

* For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index