st: Hausman test for IIA

[Kate Welti <kate.perper@gmail.com>]

Hello,

I am running the code below to try to test for IIA and it doesn't seem
to working.  Anybody have any advice?

I have a 3 level variable 0:No services 1:Clinic services 2:Non-clinic services
I want to test the null hypothesis that the inclusion of the No
Services category does not change the odds ratio of the other
pairs of choices.

xi: mlogit typeprov hismexican i.ageuslt12
est store all
xi: mlogit typeprov hismexican i.ageuslt12 if typeprov !="No services": typeprov
est store partial
hausman partial all, alleqs constant

However, I don't think the test is working as  get the following results.


Note: the rank of the differenced variance matrix (0) does not equal
the number of coefficients being tested (8); be sure
        this is what you expect, or there may be problems computing
the test.  Examine the output of your estimators for
        anything unexpected and possibly consider scaling your
variables so that the coefficients are on a similar scale.

                 ---- Coefficients ----
             |      (b)          (B)            (b-B)     sqrt(diag(V_b-V_B))
             |    partial        all         Difference          S.E.
-------------+----------------------------------------------------------------
Clinic       |
  hismexican |    .5298592     .5298592               0               0
_Iageuslt1~0 |    .7479268     .7479268               0               0
_Iageuslt1~1 |    .2958032     .2958032               0               0
       _cons |   -1.862078    -1.862078               0               0
-------------+----------------------------------------------------------------
Non_clinic   |
  hismexican |   -.1578451    -.1578451               0               0
_Iageuslt1~0 |   -.4588091    -.4588091               0               0
_Iageuslt1~1 |   -.0382422    -.0382422               0               0
       _cons |   -.6154628    -.6154628               0               0
------------------------------------------------------------------------------
                          b = consistent under Ho and Ha; obtained from mlogit
           B = inconsistent under Ha, efficient under Ho; obtained from mlogit

    Test:  Ho:  difference in coefficients not systematic

                  chi2(0) = (b-B)'[(V_b-V_B)^(-1)](b-B)
                          =        0.00
                Prob>chi2 =           .
                (V_b-V_B is not positive definite)
*
*   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/