Re: st: Constraints in mprobit

["Emily Clough" <eclough@unt.edu>]

I can get yours to work, but not mine.  Here's an example of what I've
been getting (the "Equation[1] not found" error message is the one I get
consistently):


. constraint 1 [1]ref_Chances=0

. mlogit vote_choice ref_Chances cons_Chances lib_Chances ndp_Chances,
constraints(1)

Iteration 0:   log likelihood = -2218.6871
Iteration 1:   log likelihood = -1929.9469
Iteration 2:   log likelihood =  -1902.023
Iteration 3:   log likelihood = -1900.8044
Iteration 4:   log likelihood = -1900.8035

Multinomial logistic regression                   Number of obs   =     
 1817
                                                  LR chi2(11)     =    
635.77
                                                  Prob > chi2     =    
0.0000
Log likelihood = -1900.8035                       Pseudo R2       =    
0.1433

 ( 1)  [1]ref_Chances = 0
------------------------------------------------------------------------------
 vote_choice |      Coef.   Std. Err.      z    P>|z|     [95% Conf.
Interval]
-------------+----------------------------------------------------------------
1            |
 ref_Chances |  (dropped)
cons_Chances |  -3.934843   .5529251    -7.12   0.000    -5.018557   
-2.85113
 lib_Chances |   1.350005   .4478152     3.01   0.003     .4723029   
2.227706
 ndp_Chances |   .6301302   .3626094     1.74   0.082    -.0805712   
1.340832
       _cons |  -.9765724   .1288193    -7.58   0.000    -1.229054  
-.7240913
-------------+----------------------------------------------------------------
3            |
 ref_Chances |  -.3740816    .512403    -0.73   0.465    -1.378373   
.6302099
cons_Chances |   1.469192   .5031593     2.92   0.004     .4830176   
2.455366
 lib_Chances |   1.647576   .5987774     2.75   0.006     .4739938   
2.821158
 ndp_Chances |  -3.842804   .5021989    -7.65   0.000    -4.827096  
-2.858513
       _cons |  -1.005553   .1526602    -6.59   0.000    -1.304762  
-.7063449
-------------+----------------------------------------------------------------
4            |
 ref_Chances |  -4.564323   .3240375   -14.09   0.000    -5.199425  
-3.929221
cons_Chances |   .9594781    .349486     2.75   0.006     .2744981   
1.644458
 lib_Chances |   3.817126   .3917677     9.74   0.000     3.049276   
4.584977
 ndp_Chances |   1.097337   .3428147     3.20   0.001     .4254329   
1.769242
       _cons |  -.3218146   .1107896    -2.90   0.004    -.5389582   
-.104671
------------------------------------------------------------------------------
(vote_choice==2 is the base outcome)

. mprobit vote_choice ref_Chances cons_Chances lib_Chances ndp_Chances,
constraints(1)
Constraints invalid:
equation [1] not found
r(303);




_________________________
Emily Clough, Ph.D.
Assistant Professor
Department of Political Science
University of North Texas
P. O. Box 305340
Denton, TX  76203
940-565-2214
eclough@unt.edu
>>> rgates@stata.com 05/24/06 4:45 PM >>>
Emily Clough <eclough@unt.edu> inquired about using constraints with
-mprobit-.
Here is an example:

. webuse sysdsn3
(Health insurance data)

. constraint 1 [Prepaid]age = [Uninsure]age

. mprobit insure age male nonwhite site2 site3, constraints(1)

Iteration 0:   log likelihood = -538.23525  
Iteration 1:   log likelihood = -534.84715  
Iteration 2:   log likelihood = -534.72541  
Iteration 3:   log likelihood = -534.72518  
Iteration 4:   log likelihood = -534.72518  

Multinomial probit regression                     Number of obs   =     
  615
                                                  Wald chi2(9)    =     
39.87
Log likelihood = -534.72518                       Prob > chi2     =    
0.0000

 ( 1)  [Prepaid]age - [Uninsure]age = 0
------------------------------------------------------------------------------
      insure |      Coef.   Std. Err.      z    P>|z|     [95% Conf.
Interval]
-------------+----------------------------------------------------------------
Prepaid      |
         age |  -.0086853   .0049274    -1.76   0.078    -.0183428   
.0009722
        male |   .4727575   .1716502     2.75   0.006     .1363293   
.8091858
    nonwhite |   .8230185   .1976466     4.16   0.000     .4356384   
1.210399
       site2 |    .101763   .1793036     0.57   0.570    -.2496655   
.4531915
       site3 |  -.4916245   .1903472    -2.58   0.010    -.8646981  
-.1185508
       _cons |   .1701124   .2662816     0.64   0.523      -.35179   
.6920149
-------------+----------------------------------------------------------------
Uninsure     |
         age |  -.0086853   .0049274    -1.76   0.078    -.0183428   
.0009722
        male |   .3431705   .2429219     1.41   0.158    -.1329478   
.8192887
    nonwhite |   .2643862   .2751911     0.96   0.337    -.2749785   
.8037508
       site2 |  -.6934505   .2801114    -2.48   0.013    -1.242459  
-.1444422
       site3 |  -.1884719   .2476282    -0.76   0.447    -.6738143   
.2968706
       _cons |  -.8266897   .2948186    -2.80   0.005    -1.404523  
-.2488559
------------------------------------------------------------------------------
(insure=Indemnity is the base outcome)


-Rich
rgates@stata.com
*
*   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/

*
*   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/