# Re: st: Mlogit with constraints

 From narazani To statalist@hsphsun2.harvard.edu Subject Re: st: Mlogit with constraints Date Sun, 11 Dec 2005 19:34:20 +0100

```Thank you Richard. Here you have all constraints. There are 5 alternatives and 2
independent variables.

Looking forward to hearing to you
Best, edlira

. cons 1 [2=3]: time cost1  _cons

. cons 2 [3=4]: time cost1   _cons

. cons 3 [4=5]: time cost1   _cons

. mlogit av20 time cost1  if (location>1|location<6|location>6)&purpose==1,
cons(1-3) basecat(1)

Iteration 0:   log likelihood = -479.42809
Iteration 1:   log likelihood = -259.55111
Iteration 2:   log likelihood = -256.78928
Iteration 3:   log likelihood = -256.74805
Iteration 4:   log likelihood = -256.74804

Multinomial logistic regression                   Number of obs   =        324
LR chi2(-1)     =     445.36
Prob > chi2     =          .
Log likelihood = -256.74804                       Pseudo R2       =     0.4645

( 1)  [2]time - [3]time = 0
( 2)  [2]cost1 - [3]cost1 = 0
( 3)  [2]_cons - [3]_cons = 0
( 4)  [3]time - [4]time = 0
( 5)  [3]cost1 - [4]cost1 = 0
( 6)  [3]_cons - [4]_cons = 0
( 7)  [4]time - [5]time = 0
( 8)  [4]cost1 - [5]cost1 = 0
( 9)  [4]_cons - [5]_cons = 0
------------------------------------------------------------------------------
av20 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
2            |
time |   -.015983   .0048032    -3.33   0.001    -.0253971    -.006569
cost1 |  -.0072383   .0013861    -5.22   0.000     -.009955   -.0045216
_cons |  (dropped)
-------------+----------------------------------------------------------------
3            |
time |   -.015983   .0048032    -3.33   0.001    -.0253971    -.006569
cost1 |  -.0072383   .0013861    -5.22   0.000     -.009955   -.0045216
_cons |  (dropped)
-------------+----------------------------------------------------------------
4            |
time |   -.015983   .0048032    -3.33   0.001    -.0253971    -.006569
cost1 |  -.0072383   .0013861    -5.22   0.000     -.009955   -.0045216
_cons |  (dropped)
-------------+----------------------------------------------------------------
5            |
time |   -.015983   .0048032    -3.33   0.001    -.0253971    -.006569
cost1 |  -.0072383   .0013861    -5.22   0.000     -.009955   -.0045216
_cons |  -.9755993   .3746739    -2.60   0.009    -1.709947    -.241252
------------------------------------------------------------------------------
(av20==1 is the base outcome)

sum p1-p5

Variable |       Obs        Mean    Std. Dev.       Min        Max
-------------+--------------------------------------------------------
p1 |       324    .7788082    .0914454   .5147507   .9966482
p2 |       324    .0655002    .0270792   .0009926   .1436939
p3 |       324    .0655002    .0270792   .0009926   .1436939
p4 |       324    .0655002    .0270792   .0009926   .1436939
p5 |       324    .0246914    .0102079   .0003742   .0541678

Scrive Richard Williams <Richard.A.Williams.5@ND.edu>:

> At 01:12 PM 12/11/2005, narazani wrote:
>
> >Dear all,
> >
> >I ran a multinomial logit with constraints and after that I used the command
> >"predict" to get the predicted probabilities. Stata result was: same
> >probabilities for 3 alternatives. Do you know any other command to predict
> the
> >probabilities when mlogit with constraints is used?
>
> Your results look quite peculiar.  You specify constraints 1-3 but
> mlogit reports 9 constraints.  The model has -1 degrees of
> freedom.  Most of the constants are getting dropped.
>
> I suspect something is wrong with your model specification.  We need
> to see more.  Include the actual constraints commands you are
> using.  I don't think the problem is with the predict command (which
> seems to be working correctly given the parameter estimates) but with
> the model itself.
>
> -------------------------------------------
> Richard Williams, Notre Dame Dept of Sociology
> OFFICE: (574)631-6668, (574)631-6463
> FAX:    (574)288-4373
> HOME:   (574)289-5227
> EMAIL:  Richard.A.Williams.5@ND.Edu
> WWW (personal):    http://www.nd.edu/~rwilliam
> WWW (department):    http://www.nd.edu/~soc
>
> *
> *   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/
>

----------------------------------------------------------------
This message was sent using IMP - Facolta' di Economia - Torino.

*
*   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/
```