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

[Richard Williams <Richard.A.Williams.5@ND.edu>]

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
>
> Herv� 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
WWW:    http://www.nd.edu/~rwilliam


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/