st: mlogtest hausman "invalid syntax" (Stata 8)

[nng@stanford.edu]

Hi Statalisters:


I ran a multinomial logistic regression with a 3-category dependent  
variable (the IVs include one continuous variable and 8 dummies), and  
wanted to test the IIA assumption. The problem is that -mlogtest- will  
not perform the Hausman test.  I used the -mlogtest- command after  
running -mlogit-, in different versions: I tried -mlogtest, iia-,  
-mlogtest, hausman-, and -mlogtest, smhsiao-.  Stata will do the  
Small-Hsiao test, but it won't run the -mlogtest, iia- and -mlogtest,  
hausman-  because of "invalid syntax".  I Googled generally and  
searched the Statalist archives, but have not come across anything  
directly relevant.  Am I missing something?  What might be causing this?

Code and output below -- Thanks for your attention!


. mlogit AgendaX JCSFV jcshomog ctdislctdir ctdis ctlctdir dislctdir  
courtprefF dis3 lctdir2 if term<=1999, base(2) nolog

Multinomial logistic regression                   Number of obs   =       6952
                                                  LR chi2(18)     =      85.61
                                                  Prob > chi2     =     0.0000
Log likelihood = -4858.5573                       Pseudo R2       =     0.0087

------------------------------------------------------------------------------
     AgendaX |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
CCO          |
       JCSFV |   .3233496   .1700592     1.90   0.057    -.0099604    .6566595
    jcshomog |   3.798072   1.039115     3.66   0.000     1.761443    5.834701
 ctdislctdir |   1.253344   .6129581     2.04   0.041     .0519679    2.454719
       ctdis |   .0918829   .3709508     0.25   0.804    -.6351674    .8189331
    ctlctdir |  -1.632044   .3321444    -4.91   0.000    -2.283035   -.9810527
   dislctdir |  -.4395894   .2578726    -1.70   0.088    -.9450103    .0658315
  courtprefF |   .7173647   .2016548     3.56   0.000     .3221286    1.112601
        dis3 |  -.2395791   .1785046    -1.34   0.180    -.5894416    .1102835
     lctdir2 |   .5899924   .1450551     4.07   0.000     .3056896    .8742952
       _cons |  -2.543556   .4533122    -5.61   0.000    -3.432032   -1.655081
-------------+----------------------------------------------------------------
Other        |
       JCSFV |   .0988324   .1138796     0.87   0.385    -.1243676    .3220323
    jcshomog |   1.775265   .6342595     2.80   0.005     .5321386     3.01839
 ctdislctdir |   1.428385    .381431     3.74   0.000     .6807944    2.175976
       ctdis |  -.3345637   .2452329    -1.36   0.172    -.8152114     .146084
    ctlctdir |  -.8940682   .1951554    -4.58   0.000    -1.276566   -.5115706
   dislctdir |   -.122555   .1549872    -0.79   0.429    -.4263242    .1812143
  courtprefF |   .4987686    .130574     3.82   0.000     .2428483    .7546889
        dis3 |  -.2182942   .1017204    -2.15   0.032    -.4176625   -.0189258
     lctdir2 |   .2685875   .0951733     2.82   0.005     .0820513    .4551237
       _cons |   .7143991   .2723813     2.62   0.009     .1805415    1.248257
------------------------------------------------------------------------------
(Outcome AgendaX==SSO is the comparison group)

. mlogtest, iia

**** Hausman tests of IIA assumption

 Ho: Odds(Outcome-J vs Outcome-K) are independent of other alternatives.
invalid syntax
r(110);

. mlogtest, hausman

**** Hausman tests of IIA assumption

 Ho: Odds(Outcome-J vs Outcome-K) are independent of other alternatives.
invalid syntax
r(110);

. mlogtest, smhsiao

**** Small-Hsiao tests of IIA assumption

 Ho: Odds(Outcome-J vs Outcome-K) are independent of other alternatives.

 Omitted |  lnL(full)  lnL(omit)    chi2   df   P>chi2   evidence
---------+---------------------------------------------------------
     CCO |  -1484.250  -1475.736  17.027   10    0.074   for Ho
   Other |   -532.888   -528.799   8.180   10    0.611   for Ho



Nora Ng

*
*   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/