Speed: Report margins dydx(*) at means with panel clogit

[Natan Sklair]

Hi there, I am trying to report the margins at means for a FE logit (clogit) model.

When I type "margins, dydx(*) atmeans" it never loads and I just get a spinning wheel.

Is there a way of overcoming this?

See code, below:

clogit poor_health overdebt $all_debt $L_all_debt clean_age agesq $eduvars couple bad_physical lntot_inc urban, or group(pid) robust
margins, dydx (*) atmeans

I am currently reporting odds ratios instead, but I would like to have a marginal effect for interpretation purposes.

[Joro Kolev]

You will have to provide a minimal example of data and show it with -dataex-, and then show what output containing the problem you get back from Stata.

Because with the example dataset for the command, one does not obtain "a spinning wheel."

. webuse lowbirth2, clear
(Applied Logistic Regression, Hosmer & Lemeshow)

. clogit low lwt smoke ptd ht ui i.race, group(pairid) nolog

Conditional (fixed-effects) logistic regression

                                                Number of obs     =        112
                                                LR chi2(7)        =      26.04
                                                Prob > chi2       =     0.0005
Log likelihood = -25.794271                     Pseudo R2         =     0.3355

------------------------------------------------------------------------------
         low |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         lwt |  -.0183757   .0100806    -1.82   0.068    -.0381333    .0013819
       smoke |   1.400656   .6278396     2.23   0.026     .1701131    2.631199
         ptd |   1.808009   .7886502     2.29   0.022     .2622828    3.353735
          ht |   2.361152   1.086128     2.17   0.030     .2323796    4.489924
          ui |   1.401929   .6961585     2.01   0.044     .0374836    2.766375
             |
        race |
      black  |   .5713643    .689645     0.83   0.407    -.7803149    1.923044
      other  |  -.0253148   .6992044    -0.04   0.971     -1.39573    1.345101
------------------------------------------------------------------------------

. margins, dydx (*) atmeans

Conditional marginal effects                    Number of obs     =        112
Model VCE    : OIM

Expression   : Pr(low|fixed effect is 0), predict(pu0)
dy/dx w.r.t. : lwt smoke ptd ht ui 2.race 3.race
at           : lwt             =    127.1696 (mean)
               smoke           =    .4107143 (mean)
               ptd             =    .2232143 (mean)
               ht              =    .0892857 (mean)
               ui              =    .1785714 (mean)
               1.race          =    .3928571 (mean)
               2.race          =       .1875 (mean)
               3.race          =    .4196429 (mean)

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         lwt |  -.0039303   .0009467    -4.15   0.000    -.0057858   -.0020748
       smoke |   .2995783   .2503767     1.20   0.231     -.191151    .7903075
         ptd |   .3867045   .2712309     1.43   0.154    -.1448983    .9183074
          ht |   .5050131   .2945989     1.71   0.086    -.0723902    1.082416
          ui |   .2998505   .2216245     1.35   0.176    -.1345256    .7342266
             |
        race |
      black  |   .1296423   .1709427     0.76   0.448    -.2053992    .4646838
      other  |  -.0051813   .1409953    -0.04   0.971    -.2815271    .2711645
------------------------------------------------------------------------------
Note: dy/dx for factor levels is the discrete change from the base level.

[Natan Sklair]

Joro Kolev thank you. It seems to be working now.

I initially encountered this error using xtlogit, but now with clogit it seems to be smoother