comformability error is important?

[Ayyerus Lad]

Hello to everyone!
As a stata learner, and with a lack of a supervisor, I count on generosity of cyber space. Here is my curiosity: After applying slogit, I received conformability error. I know what it means. But I'm curious about if I could ignore this message.. Because, my results are quite consistent with expectations.
Thanks in advance
Best

[Maarten Buis]

A conformability error is usually fatal, so normally you should not get results at all.

It suggest a problem deep inside the workings of Stata, so it is hard for us to diagnose what is going on exactly. The best way forward is probably to contact StataCorp directly: https://www.stata.com/support/faqs/t...-tech-support/

[Ayyerus Lad]

Thank you in fact. I'll write them. In fact there is a point that I forgot to mention, I got this message while trying to get my odds ratio by listcoef code.. If it would change anything about your answer to help me.

[daniel klein]

In fact there is a point that I forgot to mention, I got this message while trying to get my odds ratio by listcoef code..

That changes (almost) everything. listcoef is not part of official Stata but a user-written command; therefore, Stata tech-support is likely the wrong address to turn to (although they might still help you with the problem). The problem is likely not with slogit (and probably not even with the latest(!) version of listcoef).

Please do review the FAQ carefully, especially section 12. We do not have these "rules" to make it harder or more cumbersome to post questions, we have them to make it easier for everyone.

For now, make sure you have the latest version of listcoef installed, which should be

. which listcoef
...
*! version 5.0.0 2017-08-23 | long freese | stata 15 parameter naming fix

where the output already explains the likely source of your problem: changes in the naming conventions in Stata 15. I got listcoef from

spost13_ado from http://www.indiana.edu/~jslsoc/stata
    Distribution-date: 21Jan2018 / spost13_ado | SPost13 commands from Long
    and Freese (2014) / Regression Models for Categorical Outcomes using
    Stata, 3rd Edition. / Support www.indiana.edu/~jslsoc/spost.htm / Scott
    Long ([email protected]) & Jeremy Freese ([email protected])

So, install the latest version of listcoef (or run slogit and listcoef under version 14 or earlier). That should fix the problem, at least this works for me

sysuse auto , clear
slogit rep78 c.mpg i.foreign
listcoef

and yields

. sysuse auto , clear
(1978 Automobile Data)

. slogit rep78 c.mpg i.foreign

Iteration 0:   log likelihood = -141.48054  (not concave)
Iteration 1:   log likelihood = -120.22521  (not concave)
Iteration 2:   log likelihood = -96.424603  (not concave)
Iteration 3:   log likelihood = -81.449667  
Iteration 4:   log likelihood = -77.936391  
Iteration 5:   log likelihood = -77.531055  
Iteration 6:   log likelihood = -77.335904  
Iteration 7:   log likelihood = -77.277687  
Iteration 8:   log likelihood = -77.275624  
Iteration 9:   log likelihood = -77.275616  
Iteration 10:  log likelihood = -77.275616  

Stereotype logistic regression                  Number of obs     =         69
                                                Wald chi2(2)      =       2.93
Log likelihood = -77.275616                     Prob > chi2       =     0.2315

 ( 1)  [phi1_1]_cons = 1
------------------------------------------------------------------------------
       rep78 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         mpg |   .1522792   .1127969     1.35   0.177    -.0687988    .3733571
             |
     foreign |
    Foreign  |   3.827171   2.427454     1.58   0.115    -.9305516    8.584894
-------------+----------------------------------------------------------------
     /phi1_1 |          1  (constrained)
     /phi1_2 |   1.234948   .7778463     1.59   0.112    -.2896029    2.759499
     /phi1_3 |   .9229419   .5350067     1.73   0.085    -.1256519    1.971536
     /phi1_4 |   .4396286   .2900914     1.52   0.130      -.12894    1.008197
     /phi1_5 |          0  (base outcome)
-------------+----------------------------------------------------------------
     /theta1 |   3.330865   2.743523     1.21   0.225    -2.046342    8.708072
     /theta2 |    5.43129   2.507832     2.17   0.030     .5160298    10.34655
     /theta3 |   5.787148   2.247237     2.58   0.010     1.382645    10.19165
     /theta4 |   3.279139   1.735883     1.89   0.059     -.123129    6.681408
     /theta5 |          0  (base outcome)
------------------------------------------------------------------------------
(rep78=5 is the base outcome)

. listcoef

slogit (N=69): Factor change in odds

  Odds of: 5 vs 1

------------------------------------------------------------------------
             |         b        z    P>|z|       e^b   e^bStdX     SDofX
-------------+----------------------------------------------------------
         mpg |    0.1523    1.350    0.177     1.164     2.443     5.866
             |
     foreign |
    Foreign  |    3.8272    1.577    0.115    45.932     5.894     0.464
-------------+----------------------------------------------------------
phi          |
      phi1_1 |    1.0000        .        .         .         .         .
      phi1_2 |    1.2349    1.588    0.112         .         .         .
      phi1_3 |    0.9229    1.725    0.085         .         .         .
      phi1_4 |    0.4396    1.515    0.130         .         .         .
-------------+----------------------------------------------------------
theta        |
      theta1 |    3.3309    1.214    0.225         .         .         .
      theta2 |    5.4313    2.166    0.030         .         .         .
      theta3 |    5.7871    2.575    0.010         .         .         .
      theta4 |    3.2791    1.889    0.059         .         .         .
------------------------------------------------------------------------

.
end of do-file

If the problem remains or if you have any further questions, post them here.

Best
Daniel

[Ayyerus Lad]

Daniel thank you! I contacted with stata support as suggested to me here and they gave me the same answer to you. It is because of version of my command. Also, estadd listcoef doesn't produce the same error.
All my best wishes