What is meant by "ml command is unrecognized"?

[Rose Dennis]

Is the problem with Data or the command? I need to run binary logit regression. Version is Stata 15.1.

[Carlo Lazzaro]

Rose:
is your copy of Stata fully updated?

[Rose Dennis]

Yes its is.

[Nick Cox]

It's difficult to know what is going on here. You could start by telling us more about the data. Try this

egen group = group(Q9 Q13 Q14), label 

tab group

and show us the results please.

[Rose Dennis]

This is what I got:


. tab group
variable group not found
r(111);

[Nick Cox]

You didn't do what I suggested then. You need to get a new variable group before you can tabulate it.

Meanwhile #3 is alarming. Pirated versions of Stata are often corrupt and give bizarre errors, to say nothing of other aspects of using proprietary software without its being paid for.

[Rose Dennis]

Nick:
There is no variable Q13 in my data so I altered your code a little and got this result

[Nick Cox]

Sorry about mentioning Q13. Otherwise your display allows this test:

* Example generated by -dataex-. For more info, type help dataex
clear
input float freq byte(Q9 Q14 Q15)
25 0 2 1
26 0 2 2
 1 0 4 2
 1 1 2 2
83 1 3 1
25 1 3 2
10 1 4 1
 3 1 4 2
end

. logistic Q9 i.Q14 i.Q15 [fw=freq]
note: 3.Q14 != 0 predicts success perfectly;
      3.Q14 omitted and 2 obs not used.


Logistic regression                                     Number of obs =     66
                                                        LR chi2(2)    =  51.25
                                                        Prob > chi2   = 0.0000
Log likelihood = -8.4826648                             Pseudo R2     = 0.7513

------------------------------------------------------------------------------
          Q9 | Odds ratio   Std. err.      z    P>|z|     [95% conf. interval]
-------------+----------------------------------------------------------------
         Q14 |
          3  |          1  (empty)
          4  |   621.6726    901.493     4.44   0.000     36.24322    10663.43
             |
       2.Q15 |   .5933505   .8818974    -0.35   0.725     .0322233    10.92579
       _cons |   .0248879   .0290885    -3.16   0.002     .0025184    .2459559
------------------------------------------------------------------------------
Note: _cons estimates baseline odds.

.

I am presuming that Q14 and Q15 are categorical. In your case you should try

logistic Q9 i.Q14 i.Q15 

if you don't get the results above, my guess is that the copy of Stata you're using is corrupt.