fitstat

[Omer Alkan]

Why the fitstat after mprobit in stata does not work?

[Richard Williams]

Welcome to Statalist.

It could help to show your code and output using code tags. See pt 12 of the FAQ on asking questions effectively. Just saying "it does not work" doesn't give us much to go on.

It could just be that you have a syntax error. My guess, though, is that you have an outdated version of fitstat. This is the most current version:

. which fitstat
c:\ado\plus\f\fitstat.ado
*! 4.1.10 | 2017-01-05 | freese long | fixed bug in tobit (for v <14 and 15+)

If you have something different, you should type

findit spost13

and install the package, which includes fitstat and many other useful programs.

If this doesn't solve the problem, then show us your code and output using code tags.

[Richard Williams]

Also, fitstat after mprobit works fine for me.

. webuse nhanes2f, clear

. mprobit health i.female weight height, nolog

Multinomial probit regression                   Number of obs     =     10,335
                                                Wald chi2(12)     =     522.85
Log likelihood = -15490.854                     Prob > chi2       =     0.0000

------------------------------------------------------------------------------
      health |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
poor         |
    1.female |  -.3823281    .071465    -5.35   0.000    -.5223969   -.2422593
      weight |   .0037652   .0018118     2.08   0.038     .0002141    .0073162
      height |  -.0184837   .0039581    -4.67   0.000    -.0262415   -.0107258
       _cons |    2.09781   .6546169     3.20   0.001     .8147845    3.380835
-------------+----------------------------------------------------------------
fair         |
    1.female |  -.1006034   .0600426    -1.68   0.094    -.2182848     .017078
      weight |   .0060073   .0015398     3.90   0.000     .0029894    .0090252
      height |  -.0165356   .0033126    -4.99   0.000    -.0230281    -.010043
       _cons |   1.975531   .5497084     3.59   0.000     .8981228     3.05294
-------------+----------------------------------------------------------------
average      |  (base outcome)
-------------+----------------------------------------------------------------
good         |
    1.female |   .2573384   .0561469     4.58   0.000     .1472925    .3673843
      weight |  -.0057242   .0014859    -3.85   0.000    -.0086365   -.0028119
      height |   .0268221   .0030959     8.66   0.000     .0207542      .03289
       _cons |  -4.314533   .5153723    -8.37   0.000    -5.324644   -3.304422
-------------+----------------------------------------------------------------
excellent    |
    1.female |   .2208039   .0568316     3.89   0.000      .109416    .3321919
      weight |  -.0134227   .0015665    -8.57   0.000    -.0164931   -.0103523
      height |   .0425407   .0031568    13.48   0.000     .0363535    .0487279
       _cons |  -6.452372   .5235901   -12.32   0.000    -7.478589   -5.426154
------------------------------------------------------------------------------

. fitstat

                         |     mprobit 
-------------------------+-------------
Log-likelihood           |             
                   Model |  -15490.854 
-------------------------+-------------
Chi-square               |             
      Deviance(df=10319) |   30981.709 
             Wald(df=12) |     522.846 
                 p-value |       0.000 
-------------------------+-------------
R2                       |             
                   Count |       0.304 
         Count(adjusted) |       0.027 
-------------------------+-------------
IC                       |             
                     AIC |   31013.709 
        AIC divided by N |       3.001 
              BIC(df=16) |   31129.601

[Omer Alkan]

Dear Richard Williams, thanks for your help. spost13 solved the problem.