Re: st: Gologit2 - Testing whether coefficients differ

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

At 07:11 PM 12/1/2007, Kristin J. Kleinjans wrote:
> Here is in more detail what I would like to do:
> I have an outcome with 5 levels, and the brant command after ologit found
> that for the independent variable female the parallel regression
> assumption is violated. I run gologit2 allowing the coefficient on female
> to differ depending on the level of the dependent variable and get,
> therefore, four different coefficients:
> 1.871***, 1.835***, 1.480***, 1.397***.
>
> I need to formally test whether each of these four coefficients is
> different from the others. For variables which are not allowed to differ,
> I can simply do a test
> test _b[var1]==_b[var2], but I cannot figure out how to call up a
> coefficient depending on the level.
Do you want something like this?

. use "http://www.indiana.edu/~jslsoc/stata/spex_data/ordwarm2.dta";
(77 & 89 General Social Survey)

. gologit2 warm male

Generalized Ordered Logit Estimates               Number of obs   =       2293
                                                  LR chi2(3)      =      94.05
                                                  Prob > chi2     =     0.0000
Log likelihood = -2948.7475                       Pseudo R2       =     0.0157

------------------------------------------------------------------------------
        warm |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
1SD          |
        male |  -.3089732   .1246748    -2.48   0.013    -.5533313   -.0646151
       _cons |   2.057623   .0900742    22.84   0.000      1.88108    2.234165
-------------+----------------------------------------------------------------
2D           |
        male |  -.5981885    .085043    -7.03   0.000    -.7648697   -.4315073
       _cons |    .504311   .0589211     8.56   0.000     .3888278    .6197941
-------------+----------------------------------------------------------------
3A           |
        male |  -1.021626   .1194645    -8.55   0.000    -1.255772   -.7874794
       _cons |  -1.110602   .0661279   -16.79   0.000     -1.24021   -.9809933
------------------------------------------------------------------------------

. test [#1]male = [#2]male

 ( 1)  [1SD]male - [2D]male = 0

           chi2(  1) =    6.11
         Prob > chi2 =    0.0134

. test [#2]male = [#3]male

 ( 1)  [2D]male - [3A]male = 0

           chi2(  1) =   13.16
         Prob > chi2 =    0.0003

Anyway, using the equation number is the simplest way to refer to 
individual equations.


-------------------------------------------
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/