Categorical variable not found when exporting margins to LaTeX using esttab

[Jose Williams]

Hi,

I am running an ordered logit model. The outcome variable takes 5 values. When calculating the margins and exporting the results, I get an error that i.agegrp is not found. The codes that I have used:

global controls heartatk houssiz bmi diabetes i.agegrp black female    
qui ologit hlthstat $controls [pweight=finalwgt]
eststo m1: margins , dydx(*) predict(outcome(1))
qui ologit hlthstat $controls [pweight=finalwgt]
eststo m2: margins , dydx(*) predict(outcome(2))
qui ologit hlthstat $controls [pweight=finalwgt]
eststo m3: margins , dydx(*) predict(outcome(3))
qui ologit hlthstat $controls [pweight=finalwgt]
eststo m4: margins , dydx(*) predict(outcome(4))
qui ologit hlthstat $controls [pweight=finalwgt]
eststo m5: margins , dydx(*) predict(outcome(5))

esttab m1 m2 m3 m4 m5 using "omargins.tex", replace b(3) se(3) ///
 keep($controls) star(* 0.10 ** 0.05 *** 0.01) ///
 label booktabs noobs nonotes collabels(none) compress alignment(D{.}{.}{-1}) ///
 mtitle("Excellent" "V.Good" "Good" "Fair" "Poor") nobaselevels stats(N)

Example of the data:

* Example generated by -dataex-. For more info, type help dataex
clear
input byte(hlthstat heartatk houssiz) float bmi byte(diabetes agegrp black female)
2 0 4 20.495686 0 4 0 0
2 0 6 21.022337 0 3 0 1
3 0 6  24.97386 0 1 0 1
4 0 9  35.72872 1 5 0 1
2 0 3 27.923803 0 5 0 1
5 0 1 30.502144 0 5 0 1
2 0 2  23.03185 0 5 0 1
1 0 1 23.035156 0 4 0 1
2 0 3  25.63578 0 5 0 1
5 0 4  20.97999 0 5 1 0
2 0 1 31.465527 0 6 0 1
2 0 3 20.184334 0 5 1 1
3 0 4  28.54312 0 2 0 1
4 0 2   20.4144 0 1 0 1
1 0 4   24.9466 0 2 1 0
5 0 1 23.976437 1 5 1 0
5 0 7  38.44014 1 5 0 1
1 0 1 25.925926 0 6 0 1
4 0 1  20.86453 0 6 0 1
3 0 2 18.145163 0 5 0 0
2 0 2  29.09791 0 4 0 1
5 0 3  38.68736 1 4 1 0
1 0 2 21.926546 0 2 0 0
3 0 2 34.226315 0 4 0 1
4 0 2 26.902664 0 5 0 0
1 0 4  23.07461 0 2 0 0
3 0 1  24.33593 0 5 0 1
3 0 2  27.91223 0 4 0 1
3 0 3  26.32063 0 1 1 1
3 0 2  32.59102 0 5 0 0
4 0 2 25.753584 0 5 0 0
1 0 7  30.10884 0 2 1 1
1 0 1 21.084726 0 1 0 1
3 0 3   27.6745 0 5 0 0
4 0 1  22.43288 0 4 0 0
3 0 2  26.54097 0 6 0 1
2 0 4  20.00867 0 1 0 1
2 0 6 25.502905 0 1 1 1
3 0 2  26.94583 0 5 0 1
4 0 2  24.40318 0 5 0 0
2 0 3 24.241745 0 6 0 0
1 0 1  24.19344 0 1 0 0
3 0 8  28.98996 0 1 1 0
4 0 1  21.93411 0 6 0 1
3 0 1  28.57907 0 6 0 1
3 1 2 30.112375 0 5 0 0
3 0 2 20.318064 0 4 0 0
1 0 1 21.385134 0 3 0 0
5 0 2  35.72315 0 5 1 1
3 0 2 21.161545 0 2 0 0
3 0 3 22.207384 0 3 0 1
3 0 4 19.039673 0 2 0 0
1 0 3   26.5538 0 1 0 0
2 0 4  23.75451 0 1 0 0
2 0 3 28.274496 0 5 0 1
1 0 4 22.546875 0 2 0 1
3 0 5 34.772602 0 3 1 0
2 0 2  19.61235 0 5 0 0
2 0 2  30.13859 0 5 1 0
2 0 2 31.379545 0 5 0 1
1 0 1  20.84969 0 1 0 1
3 0 3   21.9425 0 3 0 0
3 0 1 20.241163 0 1 1 0
4 0 8  35.12988 0 4 1 1
3 0 2  28.21959 0 6 0 0
1 0 3 24.421795 0 4 0 0
2 0 4 21.581364 0 1 1 0
4 0 4 37.740726 1 6 0 1
3 0 2  30.77501 0 5 1 0
5 0 5 21.030283 0 6 0 1
4 1 2  25.53069 0 6 0 0
5 0 1  25.24272 0 6 0 0
2 0 2 34.823967 1 5 0 1
4 0 6 33.614716 0 5 1 1
3 0 2  27.63608 0 6 0 0
1 0 3 18.712408 0 1 1 0
3 0 2 23.012417 0 3 0 0
1 0 5  27.02887 0 3 0 0
4 0 2 26.086987 0 5 0 0
1 0 1  30.42313 0 1 0 1
2 1 2  25.67467 0 6 0 1
2 0 2  29.43503 0 5 0 1
2 0 6 25.418964 0 3 0 0
1 0 6 25.101645 0 4 0 1
3 0 4 20.557127 0 3 0 1
4 0 1  26.41101 1 6 1 0
3 0 2 26.817894 0 6 0 1
1 0 1 27.240175 0 3 0 1
4 0 4 36.435555 0 3 0 1
2 0 2  28.42347 0 2 1 1
3 0 3 28.850655 0 4 1 1
1 0 2  37.04592 0 4 0 1
3 1 1 25.452404 0 4 0 1
1 0 2 17.432068 0 6 0 0
3 0 1   31.6263 1 5 1 1
1 0 2  44.01949 0 4 0 1
2 0 2 18.321362 0 5 0 0
1 0 3  21.32432 1 1 0 1
5 0 1  18.99932 0 6 0 1
4 0 4  28.93709 1 5 0 0
end
label values hlthstat hlth
label def hlth 1 "Excellent", modify
label def hlth 2 "Very good", modify
label def hlth 3 "Good", modify
label def hlth 4 "Fair", modify
label def hlth 5 "Poor", modify
label values heartatk heartlbl
label def heartlbl 0 "No heart attack", modify
label def heartlbl 1 "Had heart attack", modify
label values diabetes diabetes
label def diabetes 0 "Not diabetic", modify
label def diabetes 1 "Diabetic", modify
label values agegrp agegrp
label def agegrp 1 "20–29", modify
label def agegrp 2 "30–39", modify
label def agegrp 3 "40–49", modify
label def agegrp 4 "50–59", modify
label def agegrp 5 "60–69", modify
label def agegrp 6 "70+", modify
label values black black
label def black 0 "Not Black", modify
label def black 1 "Black", modify
label values female female
label def female 0 "Male", modify
label def female 1 "Female", modify

[Maarten Buis]

Here is one way of doing it with the etable and collect command. This requires Stata 17, but since you did not mention which version of Stata you are using, we know that you are using Stata 17. (The default assumption on Statalist is that you have the latest version of Stata, unless you tell us otherwise)

. // estimate the model and marginal effects (note the post option for margins)
. global controls heartatk houssiz bmi diabetes i.agegrp black female    

. qui ologit hlthstat $controls

. margins , dydx(*) post

Average marginal effects                                   Number of obs = 100
Model VCE: OIM

dy/dx wrt: heartatk houssiz bmi diabetes 2.agegrp 3.agegrp 4.agegrp 5.agegrp 6.agegrp black female

1._predict: Pr(hlthstat==1), predict(pr outcome(1))
2._predict: Pr(hlthstat==2), predict(pr outcome(2))
3._predict: Pr(hlthstat==3), predict(pr outcome(3))
4._predict: Pr(hlthstat==4), predict(pr outcome(4))
5._predict: Pr(hlthstat==5), predict(pr outcome(5))

------------------------------------------------------------------------------
             |            Delta-method
             |      dy/dx   std. err.      z    P>|z|     [95% conf. interval]
-------------+----------------------------------------------------------------
heartatk     |
    _predict |
          1  |  -.0528757   .1292074    -0.41   0.682    -.3061176    .2003663
          2  |  -.0177194   .0434391    -0.41   0.683    -.1028584    .0674196
          3  |   .0138805   .0349349     0.40   0.691    -.0545906    .0823517
          4  |   .0304151   .0740619     0.41   0.681    -.1147436    .1755737
          5  |   .0262995   .0644634     0.41   0.683    -.1000465    .1526455
-------------+----------------------------------------------------------------
houssiz      |
    _predict |
          1  |  -.0241479   .0166315    -1.45   0.147    -.0567451    .0084493
          2  |  -.0080923   .0058622    -1.38   0.167    -.0195821    .0033974
          3  |   .0063391   .0050924     1.24   0.213    -.0036419    .0163201
          4  |   .0138903   .0099184     1.40   0.161    -.0055495    .0333301
          5  |   .0120108   .0087535     1.37   0.170    -.0051458    .0291673
-------------+----------------------------------------------------------------
bmi          |
    _predict |
          1  |   .0012805   .0064033     0.20   0.841    -.0112696    .0138307
          2  |   .0004291   .0021525     0.20   0.842    -.0037897     .004648
          3  |  -.0003362   .0016755    -0.20   0.841      -.00362    .0029477
          4  |  -.0007366   .0036951    -0.20   0.842    -.0079788    .0065056
          5  |  -.0006369    .003195    -0.20   0.842    -.0068991    .0056252
-------------+----------------------------------------------------------------
diabetes     |
    _predict |
          1  |  -.1536617   .1011835    -1.52   0.129    -.3519777    .0446544
          2  |  -.0514943   .0358122    -1.44   0.150    -.1216849    .0186964
          3  |   .0403381    .036976     1.09   0.275    -.0321334    .1128097
          4  |    .088389    .057326     1.54   0.123    -.0239679    .2007459
          5  |   .0764288     .04964     1.54   0.124    -.0208638    .1737214
-------------+----------------------------------------------------------------
1.agegrp     |  (base outcome)
-------------+----------------------------------------------------------------
2.agegrp     |
    _predict |
          1  |   .0565957   .1880645     0.30   0.763     -.312004    .4251954
          2  |  -.0172235    .059923    -0.29   0.774    -.1346703    .1002234
          3  |  -.0255551   .0842112    -0.30   0.762    -.1906061    .1394959
          4  |  -.0101145   .0328487    -0.31   0.758    -.0744968    .0542678
          5  |  -.0037026   .0120408    -0.31   0.758    -.0273022    .0198969
-------------+----------------------------------------------------------------
3.agegrp     |
    _predict |
          1  |   -.234376   .1502245    -1.56   0.119    -.5288106    .0600586
          2  |   .0128061   .0404654     0.32   0.752    -.0665046    .0921167
          3  |   .1186274   .0756151     1.57   0.117    -.0295756    .2668303
          4  |   .0712504   .0538957     1.32   0.186    -.0343831     .176884
          5  |   .0316922   .0284412     1.11   0.265    -.0240516    .0874359
-------------+----------------------------------------------------------------
4.agegrp     |
    _predict |
          1  |  -.2758463   .1433084    -1.92   0.054    -.5567255    .0050329
          2  |   .0000508   .0442305     0.00   0.999    -.0866394     .086741
          3  |   .1394357   .0709684     1.96   0.049     .0003402    .2785311
          4  |   .0927441   .0557569     1.66   0.096    -.0165375    .2020256
          5  |   .0436158   .0315643     1.38   0.167    -.0182491    .1054806
-------------+----------------------------------------------------------------
5.agegrp     |
    _predict |
          1  |  -.3881318   .1208778    -3.21   0.001    -.6250479   -.1512157
          2  |     -.0768   .0527458    -1.46   0.145    -.1801798    .0265799
          3  |   .1707594    .062019     2.75   0.006     .0492043    .2923144
          4  |   .1820166   .0557233     3.27   0.001      .072801    .2912322
          5  |   .1121558   .0442339     2.54   0.011      .025459    .1988527
-------------+----------------------------------------------------------------
6.agegrp     |
    _predict |
          1  |  -.4250665   .1222928    -3.48   0.001    -.6647561    -.185377
          2  |  -.1224547   .0609017    -2.01   0.044    -.2418199   -.0030896
          3  |   .1517725   .0662054     2.29   0.022     .0220124    .2815326
          4  |   .2268306    .066288     3.42   0.001     .0969086    .3567527
          5  |   .1689181   .0740365     2.28   0.023     .0238093    .3140269
-------------+----------------------------------------------------------------
black        |
    _predict |
          1  |  -.1374816   .0696338    -1.97   0.048    -.2739613   -.0010019
          2  |  -.0460721   .0265301    -1.74   0.082    -.0980701    .0059259
          3  |   .0360907   .0246947     1.46   0.144    -.0123101    .0844915
          4  |    .079082    .041495     1.91   0.057    -.0022468    .1604107
          5  |   .0683811    .039152     1.75   0.081    -.0083553    .1451176
-------------+----------------------------------------------------------------
female       |
    _predict |
          1  |   .0418097   .0571218     0.73   0.464     -.070147    .1537664
          2  |    .014011   .0193569     0.72   0.469    -.0239277    .0519498
          3  |  -.0109756   .0158667    -0.69   0.489    -.0420736    .0201225
          4  |  -.0240497   .0330139    -0.73   0.466    -.0887558    .0406564
          5  |  -.0207955   .0287448    -0.72   0.469    -.0771342    .0355432
------------------------------------------------------------------------------
Note: dy/dx for factor levels is the discrete change from the base level.

.
. // creat a table
. etable, showeq showstars showstarsnote ///
>     cstat(_r_b, nformat(%9.3f))        ///
>     cstat(_r_se, nformat(%9.3f))

---------------------------------
                            1    
---------------------------------
heartatk                         
  _predict                       
    1                   -0.053   
                       (0.129)   
    2                   -0.018   
                       (0.043)   
    3                    0.014   
                       (0.035)   
    4                    0.030   
                       (0.074)   
    5                    0.026   
                       (0.064)   
houssiz                          
  _predict                       
    1                   -0.024   
                       (0.017)   
    2                   -0.008   
                       (0.006)   
    3                    0.006   
                       (0.005)   
    4                    0.014   
                       (0.010)   
    5                    0.012   
                       (0.009)   
bmi                              
  _predict                       
    1                    0.001   
                       (0.006)   
    2                    0.000   
                       (0.002)   
    3                   -0.000   
                       (0.002)   
    4                   -0.001   
                       (0.004)   
    5                   -0.001   
                       (0.003)   
diabetes                         
  _predict                       
    1                   -0.154   
                       (0.101)   
    2                   -0.051   
                       (0.036)   
    3                    0.040   
                       (0.037)   
    4                    0.088   
                       (0.057)   
    5                    0.076   
                       (0.050)   
agegrp                           
  2                              
  _predict                       
    1                    0.057   
                       (0.188)   
    2                   -0.017   
                       (0.060)   
    3                   -0.026   
                       (0.084)   
    4                   -0.010   
                       (0.033)   
    5                   -0.004   
                       (0.012)   
  3                              
    1                   -0.234   
                       (0.150)   
    2                    0.013   
                       (0.040)   
    3                    0.119   
                       (0.076)   
    4                    0.071   
                       (0.054)   
    5                    0.032   
                       (0.028)   
  4                              
    1                   -0.276   
                       (0.143)   
    2                    0.000   
                       (0.044)   
    3                    0.139 *
                       (0.071)   
    4                    0.093   
                       (0.056)   
    5                    0.044   
                       (0.032)   
  5                              
    1                   -0.388 **
                       (0.121)   
    2                   -0.077   
                       (0.053)   
    3                    0.171 **
                       (0.062)   
    4                    0.182 **
                       (0.056)   
    5                    0.112 *
                       (0.044)   
  6                              
    1                   -0.425 **
                       (0.122)   
    2                   -0.122 *
                       (0.061)   
    3                    0.152 *
                       (0.066)   
    4                    0.227 **
                       (0.066)   
    5                    0.169 *
                       (0.074)   
black                            
  _predict                       
    1                   -0.137 *
                       (0.070)   
    2                   -0.046   
                       (0.027)   
    3                    0.036   
                       (0.025)   
    4                    0.079   
                       (0.041)   
    5                    0.068   
                       (0.039)   
female                           
  _predict                       
    1                    0.042   
                       (0.057)   
    2                    0.014   
                       (0.019)   
    3                   -0.011   
                       (0.016)   
    4                   -0.024   
                       (0.033)   
    5                   -0.021   
                       (0.029)   
Number of observations     100   
---------------------------------
** p<.01, * p<.05

.
. // reorder such that the predictions are the columns
. collect layout (coleq#result[_r_b _r_se]) ///
>                 (colname[]#stars)

Collection: ETable
      Rows: coleq#result[_r_b _r_se]
   Columns: colname[]#stars
   Table 1: 23 x 10

-------------------------------------------------------------
                               _predict                      
              1         2          3          4         5    
-------------------------------------------------------------
heartatk  -0.053     -0.018     0.014      0.030      0.026  
         (0.129)    (0.043)   (0.035)    (0.074)    (0.064)  
houssiz   -0.024     -0.008     0.006      0.014      0.012  
         (0.017)    (0.006)   (0.005)    (0.010)    (0.009)  
bmi        0.001      0.000    -0.000     -0.001     -0.001  
         (0.006)    (0.002)   (0.002)    (0.004)    (0.003)  
diabetes  -0.154     -0.051     0.040      0.088      0.076  
         (0.101)    (0.036)   (0.037)    (0.057)    (0.050)  
agegrp                                                       
  2        0.057     -0.017    -0.026     -0.010     -0.004  
         (0.188)    (0.060)   (0.084)    (0.033)    (0.012)  
  3       -0.234      0.013     0.119      0.071      0.032  
         (0.150)    (0.040)   (0.076)    (0.054)    (0.028)  
  4       -0.276      0.000     0.139 *    0.093      0.044  
         (0.143)    (0.044)   (0.071)    (0.056)    (0.032)  
  5       -0.388 **  -0.077     0.171 **   0.182 **   0.112 *
         (0.121)    (0.053)   (0.062)    (0.056)    (0.044)  
  6       -0.425 **  -0.122 *   0.152 *    0.227 **   0.169 *
         (0.122)    (0.061)   (0.066)    (0.066)    (0.074)  
black     -0.137 *   -0.046     0.036      0.079      0.068  
         (0.070)    (0.027)   (0.025)    (0.041)    (0.039)  
female     0.042      0.014    -0.011     -0.024     -0.021  
         (0.057)    (0.019)   (0.016)    (0.033)    (0.029)  
-------------------------------------------------------------
** p<.01, * p<.05

.
. // give age and predictions labels                
. collect label levels coleq    ///
>     2.agegrp "30-39"          ///
>     3.agegrp "40-49"          ///
>     4.agegrp "50-59"          ///
>     5.agegrp "60-69"          ///
>     6.agegrp "70+"

.     
. collect label levels colname  ///
>     1._predict "Excellent"    ///
>     2._predict "V.Good"       ///
>     3._predict "Good"         ///
>     4._predict "Fair"         ///
>     5._predict "Poor"             

.
. // admire the result    
. collect preview

-------------------------------------------------------------
          Excellent   V.Good     Good       Fair       Poor  
                                                             
-------------------------------------------------------------
heartatk  -0.053     -0.018     0.014      0.030      0.026  
         (0.129)    (0.043)   (0.035)    (0.074)    (0.064)  
houssiz   -0.024     -0.008     0.006      0.014      0.012  
         (0.017)    (0.006)   (0.005)    (0.010)    (0.009)  
bmi        0.001      0.000    -0.000     -0.001     -0.001  
         (0.006)    (0.002)   (0.002)    (0.004)    (0.003)  
diabetes  -0.154     -0.051     0.040      0.088      0.076  
         (0.101)    (0.036)   (0.037)    (0.057)    (0.050)  
30-39      0.057     -0.017    -0.026     -0.010     -0.004  
         (0.188)    (0.060)   (0.084)    (0.033)    (0.012)  
40-49     -0.234      0.013     0.119      0.071      0.032  
         (0.150)    (0.040)   (0.076)    (0.054)    (0.028)  
50-59     -0.276      0.000     0.139 *    0.093      0.044  
         (0.143)    (0.044)   (0.071)    (0.056)    (0.032)  
60-69     -0.388 **  -0.077     0.171 **   0.182 **   0.112 *
         (0.121)    (0.053)   (0.062)    (0.056)    (0.044)  
70+       -0.425 **  -0.122 *   0.152 *    0.227 **   0.169 *
         (0.122)    (0.061)   (0.066)    (0.066)    (0.074)  
black     -0.137 *   -0.046     0.036      0.079      0.068  
         (0.070)    (0.027)   (0.025)    (0.041)    (0.039)  
female     0.042      0.014    -0.011     -0.024     -0.021  
         (0.057)    (0.019)   (0.016)    (0.033)    (0.029)  
-------------------------------------------------------------
** p<.01, * p<.05

.
. // export to LaTeX
. // (I like it without \begin{table} so I can add \label{})
. collect style tex , nobegintable

. collect export c:\temp\foo.tex, replace tableonly
(collection ETable exported to file c:\temp\foo.tex)