margins with interaction term

[Rasmus Madsen]

Hello everyone,

I am having troubles getting my margins with following commands (it just gives me "not estimable"):

reg log_co2_capita c.log_gdp_capita##c.log_gdp_capita##i.dum1986 oil_country oil_gdp_capita log_num_years_demo i.country i.year, vce(robust)

summarize log_gdp_capita if electoral_2==1
global MyMin = r(min)
global MyMax = r(max)
global Grid = (15-$MyMin)/100

margins, dydx(c.log_gdp_capita) at(c.log_gdp_capita=($MyMin($Grid)15) dum1986=(0))

I want to get so I can plot two marginsplot, one where the margins are with dum1986 = 0 and a plot with dum1986 = 1.

The code works just fine if I am not trying to split the margins upon the dum1986 but just running following: margins, dydx(c.log_gdp_capita) at(c.log_gdp_capita=($MyMin($Grid)15))

How can I solve this?

Best Rasmus

[Clyde Schechter]

You are more likely to get a timely and helpful response if you provide more information. There is nothing obviously wrong with the code, and with "not estimable" problems, the trouble is almost always with the model or the data (or both). I suggest you post back showing, a) example data that includes all variable necessary to run the regression command, b) the complete output you get from the -regress- command (including all messages that appear before the regression table itself), and c) the output of the -margins- command, again including any messages there might be.

So that the example data will be helpful for troubleshooting, you should use the -dataex- command to post it. Also, you should run the regression and margins commands on the example data to verify that it reproduces the problem you are having.

If you are running version 18, 17, 16 or a fully updated version 15.1 or 14.2, -dataex- is already part of your official Stata installation. If not, run -ssc install dataex- to get it. Either way, run -help dataex- to read the simple instructions for using it. -dataex- will save you time; it is easier and quicker than typing out tables. It includes complete information about aspects of the data that are often critical to answering your question but cannot be seen from tabular displays or screenshots. It also makes it possible for those who want to help you to create a faithful representation of your example to try out their code, which in turn makes it more likely that their answer will actually work in your data.

[Rasmus Madsen]

You are more likely to get a timely and helpful response if you provide more information. There is nothing obviously wrong with the code, and with "not estimable" problems, the trouble is almost always with the model or the data (or both). I suggest you post back showing, a) example data that includes all variable necessary to run the regression command, b) the complete output you get from the -regress- command (including all messages that appear before the regression table itself), and c) the output of the -margins- command, again including any messages there might be.

So that the example data will be helpful for troubleshooting, you should use the -dataex- command to post it. Also, you should run the regression and margins commands on the example data to verify that it reproduces the problem you are having.

If you are running version 18, 17, 16 or a fully updated version 15.1 or 14.2, -dataex- is already part of your official Stata installation. If not, run -ssc install dataex- to get it. Either way, run -help dataex- to read the simple instructions for using it. -dataex- will save you time; it is easier and quicker than typing out tables. It includes complete information about aspects of the data that are often critical to answering your question but cannot be seen from tabular displays or screenshots. It also makes it possible for those who want to help you to create a faithful representation of your example to try out their code, which in turn makes it more likely that their answer will actually work in your data.

Thanks for reply.

A)
. dataex year log_co2_capita log_gdp_capita oil_gdp_capita oil_country log_num_years_demo dum1986 id, count(200)

----------------------- copy starting from the next line -----------------------

* Example generated by -dataex-. For more info, type help dataex
clear
input int year float(log_co2_capita log_gdp_capita oil_gdp_capita oil_country log_num_years_demo dum1986 id)
1950 -4.4860005  7.052721         0 0         0 0 1
1951 -4.4147205  7.064759         0 0         0 0 1
1952  -4.427319  7.080868         0 0         0 0 1
1953 -4.2914724  7.122867         0 0         0 0 1
1954  -4.304236  7.126891         0 0         0 0 1
1955  -3.947457  7.127694         0 0         0 0 1
1956  -3.787525  7.153052         0 0         0 0 1
1957 -3.3325515  7.133296         0 0         0 0 1
1958  -3.229713   7.16858         0 0         0 0 1
1959  -3.091947   7.17549         0 0         0 0 1
1960  -3.036539  7.189922         0 0         0 0 1
1961  -2.885353  7.177019         0 0         0 0 1
1962  -2.566884  7.171657         0 0         0 0 1
1963  -2.561667   7.16858         0 0         0 0 1
1964  -2.412045  7.163172         0 0         0 0 1
1965 -2.2512336  7.162397         0 0         0 0 1
1966 -2.1934214  7.148346         0 0         0 0 1
1967 -2.0552723  7.152269         0 0         0 0 1
1968 -2.1254349  7.162397         0 0         0 0 1
1969  -2.411416  7.153052         0 0         0 0 1
1970  -1.862121  7.148346         0 0         0 0 1
1971   -1.76088  7.120444         0 0         0 0 1
1972 -1.9981352  6.914731         0 0         0 0 1
1973 -1.9569538  6.918695         0 0         0 0 1
1974  -1.825252  6.946014         0 0         0 0 1
1975  -1.745869  6.979145         0 0         0 0 1
1976  -1.836202  7.007601         0 0         0 0 1
1977  -1.671749  6.929517         0 0         0 0 1
1978  -1.793234  6.975414         0 0         0 0 1
1979  -1.760665  6.930495         0 0         0 0 1
1980 -1.9614488  6.926577         0 0         0 0 1
1981  -1.729585  7.042286         0 0         0 0 1
1982  -1.572022  7.146772         0 0         0 0 1
1983 -1.3734773  7.205635         0 0         0 0 1
1984  -1.289379  7.198184         0 0         0 0 1
1985 -1.0993692  7.173192         0 0         0 0 1
1986 -1.2042565  7.203405         0 0         0 1 1
1987 -1.1984997  7.099202         0 0         0 1 1
1988 -1.2904783  7.003974         0 0         0 1 1
1989  -1.350745  6.906755         0 0         0 1 1
1990 -1.6645212  6.870053         0 0         0 1 1
1991 -1.7251036  6.781251         0 0         0 1 1
1992 -2.0962527  6.738005         0 0         0 1 1
1993 -2.2426038  6.360271         0 0         0 1 1
1994  -2.363768  6.060115         0 0         0 1 1
1995  -2.449661  6.450376         0 0         0 1 1
1996  -2.524578  6.397222         0 0         0 1 1
1997  -2.613017  6.346685         0 0         0 1 1
1998  -2.671709  6.300857         0 0         0 1 1
1999 -2.8704975  6.251245         0 0         0 1 1
2000 -2.9265654  6.219342         0 0         0 1 1
2001  -2.913226  6.193756         0 0         0 1 1
2002  -2.751123  6.680624         0 0         0 1 1
2003 -2.6755145  6.736736         0 0         0 1 1
2004  -2.946388  6.767388         0 0         0 1 1
2005  -2.558726  6.871514         0 0         0 1 1
2006  -2.466646  6.963281         0 0         0 1 1
2007 -2.2247837  7.138865         0 0         0 1 1
2008  -1.826422   7.18509         0 0         0 1 1
2009  -1.454977  7.350722         0 0         0 1 1
2010  -1.214923  7.394906         0 0         0 1 1
2011  -.9045094  7.491087         0 0         0 1 1
2012 -1.1105169  7.573017         0 0         0 1 1
2013 -1.2266164  7.613325         0 0         0 1 1
2014 -1.2718998  7.611843         0 0         0 1 1
2015  -1.237611  7.564239         0 0         0 1 1
2016 -1.3401924  7.564757         0 0         0 1 1
2017 -1.2842704  7.608248         0 0         0 1 1
2018  -1.221201  7.567633         0 0         0 1 1
2019 -1.2262138  7.549066         0 0         0 1 1
2020 -1.1873144  7.493927         0 0         0 1 1
2021 -1.1831347  7.232991         0 0         0 1 1
1950 -1.4401605  7.375256         0 0         0 0 2
1951 -1.1628684  7.418181         0 0         0 0 2
1952 -1.2672733  7.418781         0 0         0 0 2
1953  -1.194388  7.459339         0 0         0 0 2
1954 -1.0322244  7.487174         0 0         0 0 2
1955  -.7851053   7.54009         0 0         0 0 2
1956 -.58167773  7.550661         0 0         0 0 2
1957 -.02878458  7.612337         0 0         0 0 2
1958   -.287078  7.656337         0 0         0 0 2
1959 -.13686895  7.696667         0 0         0 0 2
1960  .16701297  7.746301         0 0         0 0 2
1961   .2530744  7.754481         0 0         0 0 2
1962   .2975151  7.786967         0 0         0 0 2
1963   .0969865  7.820439         0 0         0 0 2
1964 .033141118  7.853993         0 0         0 0 2
1965  .07768985  7.889834         0 0         0 0 2
1966  .20804445  7.926602         0 0         0 0 2
1967  .22761075  7.964156         0 0         0 0 2
1968  .33502355  7.999343         0 0         0 0 2
1969   .3607004  8.032685         0 0         0 0 2
1970    .475343  8.069029         0 0         0 0 2
1971   .5985032  8.108322         0 0         0 0 2
1972   .8313248   8.14642         0 0         0 0 2
1973   .7407044  8.195058         0 0         0 0 2
1974   .5183203  8.198915         0 0         0 0 2
1975  .54952735 8.2022085         0 0         0 0 2
1976   .6005585  8.206584         0 0         0 0 2
1977   .7229262  8.210939         0 0         0 0 2
1978   .8290063  8.215007         0 0         0 0 2
1979   .9652082  8.220403         0 0         0 0 2
1980   .5640639  8.227108         0 0         0 0 2
1981   .8972146  8.232706         0 0         0 0 2
1982   .8766676  8.238273         0 0         0 0 2
1983   .9047014  8.244071         0 0         0 0 2
1984   .9158533  8.249576         0 0         0 0 2
1985   .9102471  8.254789         0 0         0 0 2
1986   .9212987   8.26101         0 0         0 1 2
1987    .832966  8.267192         0 0         0 1 2
1988   .8092165  8.273847         0 0         0 1 2
1989  1.0071766  8.280965         0 0         0 1 2
1990  .51605856   8.28979         0 0         0 1 2
1991   .2617847  7.986879         0 0         0 1 2
1992 -.27194077  7.952181         0 0         0 1 2
1993  -.3448437   8.05103         0 0  .6931472 1 2
1994 -.53779745  8.131446         0 0 1.0986123 1 2
1995  -.4524581  8.207034         0 0 1.3862944 1 2
1996  -.4833354  8.285434         0 0         0 1 2
1997  -.7460648  8.168199         0 0         0 1 2
1998  -.6108554  8.261881         0 0  .6931472 1 2
1999 -.07172997  8.396616         0 0 1.0986123 1 2
2000 -.05063179  8.478136         0 0 1.3862944 1 2
2001  .02103569  8.572781         0 0  1.609438 1 2
2002  .18232347  8.632121         0 0 1.7917595 1 2
2003  .33028916  8.700289         0 0   1.94591 1 2
2004   .3102439  8.767153         0 0 2.0794415 1 2
2005   .3401686   8.83324         0 0 2.1972246 1 2
2006  .26397818  8.902092         0 0 2.3025851 1 2
2007  .28284478  8.970325         0 0  2.397895 1 2
2008   .3985544  9.050422         0 0  2.484907 1 2
2009   .4079014  9.089242         0 0  2.564949 1 2
2010   .4959273  9.129453         0 0 2.6390574 1 2
2011   .6055338  9.157361         0 0   2.70805 1 2
2012   .5169767  9.168685         0 0  2.772589 1 2
2013   .6051145  9.175749         0 0  2.833213 1 2
2014   .7324882  9.190953         0 0  2.890372 1 2
2015   .4914913  9.213535         0 0  2.944439 1 2
2016   .4748245  9.243969         0 0  2.995732 1 2
2017   .6088259  9.278197         0 0 3.0445225 1 2
2018     .53145  9.315076         0 0 3.0910425 1 2
2019   .5185495  9.430714         0 0  3.135494 1 2
2020  .56000113  9.402877         0 0  3.178054 1 2
2021     .54101  9.497484         0 0  3.218876 1 2
2022   .5557227  9.557064         0 0 3.2580965 1 2
1950  -.8687522  7.685244  7.685244 1         0 0 3
1951  -.8084084  7.671827  7.671827 1         0 0 3
1952  -.8953685  7.693026  7.693026 1         0 0 3
1953  -.8934432  7.687997  7.687997 1         0 0 3
1954  -.8800764  7.736743  7.736743 1         0 0 3
1955  -.7990406  7.741968  7.741968 1         0 0 3
1956  -.7403684  7.813996  7.813996 1         0 0 3
1957  -.6590798  7.900637  7.900637 1         0 0 3
1958  -.7404877  7.915713  7.915713 1         0 0 3
1959  -.6798379  8.064008  8.064008 1         0 0 3
1960  -.6164603 8.1101265 8.1101265 1         0 0 3
1961  -.6498494   7.96137   7.96137 1         0 0 3
1962  -.7326536  7.733684  7.733684 1         0 0 3
1963  -.7924178  7.943783  7.943783 1         0 0 3
1964  -.7692975  7.965199  7.965199 1         0 0 3
1965  -.6308535  8.000014  8.000014 1         0 0 3
1966  -.4040474  7.919356  7.919356 1         0 0 3
1967  -.4250191  7.974877  7.974877 1         0 0 3
1968  -.3767424  8.055475  8.055475 1         0 0 3
1969  -.1805043  8.118207  8.118207 1         0 0 3
1970  .08760063  8.184514  8.184514 1         0 0 3
1971  .27891812  8.067149  8.067149 1         0 0 3
1972   .6736767  8.228444  8.228444 1         0 0 3
1973    .951147  8.231376  8.231376 1         0 0 3
1974   .7436359   8.26101   8.26101 1         0 0 3
1975    .710541  8.299037  8.299037 1         0 0 3
1976   .8635251  8.332549  8.332549 1         0 0 3
1977    .892894  8.388906  8.388906 1         0 0 3
1978  1.2646984 8.4788685 8.4788685 1         0 0 3
1979   .9191071   8.53464   8.53464 1         0 0 3
1980   1.265318  8.521981  8.521981 1         0 0 3
1981   .8732858  8.515391  8.515391 1         0 0 3
1982   .6722514   8.54578   8.54578 1         0 0 3
1983   .9316186   8.56465   8.56465 1         0 0 3
1984   1.198838 8.5869055 8.5869055 1         0 0 3
1985  1.1880243  8.606851  8.606851 1         0 0 3
1986  1.2014155  8.568267  8.568267 1         0 1 3
1987  1.2687714   8.53464   8.53464 1         0 1 3
1988  1.2392062  8.486734  8.486734 1         0 1 3
1989  1.1651803  8.494743  8.494743 1         0 1 3
1990  1.1010051 8.4546795 8.4546795 1         0 1 3
1991  1.1037662 8.4570675 8.4570675 1         0 1 3
1992  1.0940961   8.48678   8.48678 1         0 1 3
1993   1.097161  8.479891  8.479891 1         0 1 3
1994  1.1542745 8.4881735 8.4881735 1         0 1 3
1995  1.2232758  8.545272  8.545272 1         0 1 3
1996  1.2387984  8.603671  8.603671 1         0 1 3
1997  1.0979486   8.63737   8.63737 1         0 1 3
1998  1.2698052  8.711077  8.711077 1         0 1 3
1999   1.103115  8.767408  8.767408 1         0 1 3
2000  1.0206393  8.829746  8.829746 1         0 1 3
2001   1.023609  8.884367  8.884367 1         0 1 3
2002   1.053535  8.963779  8.963779 1         0 1 3
2003  1.0819719  9.058374  9.058374 1         0 1 3
2004   1.037124  9.125032  9.125032 1         0 1 3
end

------------------ copy up to and including the previous line ------------------

Listed 200 out of 10470 observations


B)
. reg log_co2_capita c.log_gdp_capita##c.log_gdp_capita##i.dum1986 oil_country oil_gdp_capita log_num_years_demo i.id i.year if electoral_2 == 1, vce(robust)
note: log_num_years_demo omitted because of collinearity.
note: 150.id omitted because of collinearity.
note: 2022.year omitted because of collinearity.

Linear regression Number of obs = 3,088
F(189, 2896) = .
Prob > F = .
R-squared = 0.9453
Root MSE = .49498

-----------------------------------------------------------------------------------------------------------
| Robust
log_co2_capita | Coefficient std. err. t P>|t| [95% conf. interval]
------------------------------------------+----------------------------------------------------------------
log_gdp_capita | 2.447508 .3860021 6.34 0.000 1.690642 3.204375
|
c.log_gdp_capita#c.log_gdp_capita | -.1051254 .0231573 -4.54 0.000 -.1505317 -.059719
|
1.dum1986 | 2.417892 1.49325 1.62 0.106 -.5100487 5.345832
|
dum1986#c.log_gdp_capita |
1 | -.3123923 .3750778 -0.83 0.405 -1.047839 .423054
|
dum1986#c.log_gdp_capita#c.log_gdp_capita |
1 | .0240187 .0233163 1.03 0.303 -.0216995 .0697368
|
oil_country | 1.725817 .7411381 2.33 0.020 .2726058 3.179028
oil_gdp_capita | .1332774 .0780715 1.71 0.088 -.0198039 .2863588
log_num_years_demo | 0 (omitted)
|
id |
2 | 1.648538 .1215444 13.56 0.000 1.410216 1.886861
3 | -1.351147 .2437722 -5.54 0.000 -1.829132 -.8731627
4 | -1.9077 .3141311 -6.07 0.000 -2.523643 -1.291757
5 | 1.596562 .1711483 9.33 0.000 1.260977 1.932147
6 | 1.067704 .1470488 7.26 0.000 .7793737 1.356035
9 | 2.223471 .1413519 15.73 0.000 1.946311 2.500632
10 | 3.367411 .1560406 21.58 0.000 3.06145 3.673373
11 | -.4235637 .0930608 -4.55 0.000 -.6060358 -.2410915
12 | 2.437219 .1492637 16.33 0.000 2.144545 2.729892
14 | -.7732361 .0932373 -8.29 0.000 -.9560543 -.590418
15 | .513396 .131199 3.91 0.000 .2561432 .7706487
16 | 1.93069 .1422626 13.57 0.000 1.651743 2.209636
18 | -1.765334 .2719834 -6.49 0.000 -2.298635 -1.232034
19 | 2.457432 .1502158 16.36 0.000 2.162891 2.751972
20 | -1.39391 .1027819 -13.56 0.000 -1.595444 -1.192377
21 | -1.503362 .1000558 -15.03 0.000 -1.69955 -1.307174
22 | -1.110661 .1756536 -6.32 0.000 -1.45508 -.7662425
23 | -.0573685 .1406798 -0.41 0.683 -.3332112 .2184742
25 | .4353536 .1703661 2.56 0.011 .1013026 .7694047
26 | -.8899644 .0950735 -9.36 0.000 -1.076383 -.703546
27 | -.8229522 .0967376 -8.51 0.000 -1.012634 -.6332706
28 | 1.028322 .1492506 6.89 0.000 .7356734 1.32097
29 | -.9523986 .2610571 -3.65 0.000 -1.464275 -.440522
31 | -.0973981 .1113515 -0.87 0.382 -.3157343 .120938
32 | .4236528 .1285993 3.29 0.001 .1714975 .6758081
34 | .3041823 .1116403 2.72 0.006 .0852799 .5230847
35 | 1.491855 .1560842 9.56 0.000 1.185808 1.797903
36 | 1.716017 .1233059 13.92 0.000 1.47424 1.957793
37 | 1.736282 .1602546 10.83 0.000 1.422057 2.050506
38 | 3.01656 .1583897 19.05 0.000 2.705992 3.327128
39 | -.0288111 .0818987 -0.35 0.725 -.1893967 .1317745
41 | .6744261 .1018897 6.62 0.000 .4746425 .8742097
42 | .9241054 .1234376 7.49 0.000 .6820709 1.16614
43 | .5285162 .1362006 3.88 0.000 .2614564 .7955761
44 | 1.314244 .1253197 10.49 0.000 1.068519 1.559969
45 | .1864795 .1215504 1.53 0.125 -.0518546 .4248136
46 | .2379298 .178308 1.33 0.182 -.1116936 .5875531
47 | 3.066098 .1664605 18.42 0.000 2.739705 3.392491
48 | -1.092014 .1001097 -10.91 0.000 -1.288308 -.8957208
51 | 2.210651 .1525905 14.49 0.000 1.911454 2.509848
53 | 1.06066 .1556781 6.81 0.000 .7554089 1.365911
55 | .0441718 .0975721 0.45 0.651 -.1471459 .2354895
56 | 1.593117 .1631522 9.76 0.000 1.273211 1.913023
57 | -.2854558 .1308733 -2.18 0.029 -.54207 -.0288416
58 | .5744065 .0905277 6.35 0.000 .3969013 .7519116
59 | -.2459374 .102417 -2.40 0.016 -.446755 -.0451197
60 | -.6811488 .1054293 -6.46 0.000 -.8878728 -.4744248
61 | .4430737 .1124022 3.94 0.000 .2226774 .66347
62 | 1.276167 .1625687 7.85 0.000 .9574049 1.594929
63 | 2.498227 .1597755 15.64 0.000 2.184942 2.811513
66 | .7180703 .1129269 6.36 0.000 .496645 .9394956
67 | -1.140505 .2379792 -4.79 0.000 -1.607131 -.6738794
68 | -1.195267 .2323351 -5.14 0.000 -1.650826 -.7397079
74 | 1.226601 .1344595 9.12 0.000 .9629548 1.490247
75 | 2.733816 .1529407 17.87 0.000 2.433932 3.033699
76 | .3008729 .1045182 2.88 0.004 .0959352 .5058106
77 | -.4135611 .2914684 -1.42 0.156 -.9850675 .1579454
78 | 2.087349 .1696158 12.31 0.000 1.754769 2.419929
79 | 1.47854 .1278396 11.57 0.000 1.227874 1.729206
80 | 1.316887 .1762688 7.47 0.000 .9712617 1.662512
81 | .1831241 .1558692 1.17 0.240 -.1225016 .4887498
82 | -.5198649 .2395862 -2.17 0.030 -.9896416 -.0500882
83 | 2.29192 .161474 14.19 0.000 1.975305 2.608536
85 | -.5716093 .0959483 -5.96 0.000 -.7597432 -.3834755
86 | -.1994278 .0994415 -2.01 0.045 -.3944111 -.0044444
87 | 1.84681 .2065761 8.94 0.000 1.441759 2.251861
88 | -1.007024 .0893089 -11.28 0.000 -1.182139 -.8319086
90 | .6890059 .1245368 5.53 0.000 .4448162 .9331956
93 | 2.320207 .1472744 15.75 0.000 2.031434 2.60898
94 | 2.804123 .0898839 31.20 0.000 2.62788 2.980366
95 | 1.637082 .1356367 12.07 0.000 1.371128 1.903037
96 | .8036897 .1300296 6.18 0.000 .5487297 1.05865
97 | -.4748767 .1690771 -2.81 0.005 -.8064004 -.143353
98 | -.1178081 .0940065 -1.25 0.210 -.3021345 .0665182
100 | -1.76455 .2074507 -8.51 0.000 -2.171316 -1.357784
103 | .3077333 .1149496 2.68 0.007 .082342 .5331246
104 | -.8181719 .1239758 -6.60 0.000 -1.061262 -.5750823
105 | -1.782568 .2808912 -6.35 0.000 -2.333335 -1.231801
106 | 2.367688 .113126 20.93 0.000 2.145873 2.589504
107 | 1.992505 .146559 13.60 0.000 1.705135 2.279876
109 | -1.448679 .2940058 -4.93 0.000 -2.02516 -.8721969
110 | .8071738 .1167611 6.91 0.000 .5782305 1.036117
111 | 1.199527 .1413214 8.49 0.000 .9224258 1.476627
113 | .8802413 .1398337 6.29 0.000 .6060577 1.154425
115 | 2.947206 .1554217 18.96 0.000 2.642458 3.251954
117 | .1878428 .2470651 0.76 0.447 -.2965983 .6722839
118 | 3.003141 .1251068 24.00 0.000 2.757834 3.248448
119 | -.0273842 .223397 -0.12 0.902 -.4654175 .410649
120 | -1.225804 .1235428 -9.92 0.000 -1.468045 -.9835634
121 | -.970289 .2420793 -4.01 0.000 -1.444954 -.495624
122 | .1410983 .2042969 0.69 0.490 -.2594837 .5416804
123 | 2.171202 .1468796 14.78 0.000 1.883202 2.459201
124 | -.2995944 .1308289 -2.29 0.022 -.5561215 -.0430672
126 | 2.415 .155141 15.57 0.000 2.110802 2.719197
127 | 1.962413 .1712615 11.46 0.000 1.626606 2.298219
130 | 1.843307 .1627086 11.33 0.000 1.52427 2.162343
131 | -.6309087 .1167208 -5.41 0.000 -.859773 -.4020444
134 | .8623093 .1499973 5.75 0.000 .568197 1.156422
135 | 1.356536 .1422084 9.54 0.000 1.077696 1.635376
136 | .106755 .097871 1.09 0.275 -.0851488 .2986587
137 | .7934478 .1503094 5.28 0.000 .4987236 1.088172
138 | -.0234167 .1090755 -0.21 0.830 -.2372901 .1904567
140 | 1.034039 .1227703 8.42 0.000 .7933127 1.274765
141 | 1.109223 .1503654 7.38 0.000 .8143888 1.404057
142 | 2.620819 .1310139 20.00 0.000 2.363929 2.877709
143 | -.8115912 .0956225 -8.49 0.000 -.9990862 -.6240963
144 | 2.808496 .1449996 19.37 0.000 2.524183 3.092809
145 | -.887888 .3199692 -2.77 0.006 -1.515278 -.2604976
148 | .8738072 .1629157 5.36 0.000 .5543648 1.193249
149 | 2.012543 .1370214 14.69 0.000 1.743873 2.281212
150 | 0 (omitted)
151 | .739769 .1088493 6.80 0.000 .5263391 .9531989
152 | .558662 .1716756 3.25 0.001 .2220433 .8952807
153 | 1.02018 .1012807 10.07 0.000 .8215906 1.21877
|
year |
1951 | -.104174 .2177826 -0.48 0.632 -.5311984 .3228505
1952 | .1609452 .1829899 0.88 0.379 -.1978585 .5197488
1953 | .1181091 .1790174 0.66 0.509 -.2329052 .4691235
1954 | .3282822 .1675145 1.96 0.050 -.0001774 .6567419
1955 | .3542887 .1619655 2.19 0.029 .0367095 .6718679
1956 | .394283 .1674571 2.35 0.019 .0659359 .7226301
1957 | .4288074 .1633845 2.62 0.009 .1084456 .7491691
1958 | .4339267 .1670798 2.60 0.009 .1063193 .7615341
1959 | .4016474 .1961776 2.05 0.041 .0169855 .7863093
1960 | .4473872 .1955958 2.29 0.022 .0638661 .8309083
1961 | .4838388 .1908222 2.54 0.011 .1096778 .8579998
1962 | .5302446 .1834892 2.89 0.004 .170462 .8900272
1963 | .6117313 .1842562 3.32 0.001 .2504447 .9730179
1964 | .6047214 .1946389 3.11 0.002 .2230767 .9863661
1965 | .6476553 .186653 3.47 0.001 .2816692 1.013641
1966 | .6514841 .1805471 3.61 0.000 .2974703 1.005498
1967 | .7282828 .1568411 4.64 0.000 .4207514 1.035814
1968 | .7844765 .1657474 4.73 0.000 .4594818 1.109471
1969 | .9080472 .1650974 5.50 0.000 .5843268 1.231768
1970 | 1.015652 .1618504 6.28 0.000 .698298 1.333005
1971 | 1.055495 .1571011 6.72 0.000 .7474538 1.363536
1972 | 1.04451 .1574125 6.64 0.000 .7358585 1.353162
1973 | 1.126949 .1579138 7.14 0.000 .8173138 1.436583
1974 | 1.095542 .1592347 6.88 0.000 .7833171 1.407767
1975 | 1.128957 .1585563 7.12 0.000 .8180624 1.439852
1976 | 1.126139 .1582136 7.12 0.000 .8159166 1.436362
1977 | 1.116516 .1585513 7.04 0.000 .8056307 1.4274
1978 | 1.145718 .1612799 7.10 0.000 .829483 1.461953
1979 | 1.174632 .1644516 7.14 0.000 .8521775 1.497086
1980 | 1.235588 .1583881 7.80 0.000 .925023 1.546153
1981 | 1.214034 .1585018 7.66 0.000 .9032467 1.524822
1982 | 1.209616 .157564 7.68 0.000 .9006676 1.518565
1983 | 1.213399 .1583138 7.66 0.000 .9029802 1.523818
1984 | 1.226054 .1577775 7.77 0.000 .9166864 1.535422
1985 | 1.252878 .1582265 7.92 0.000 .94263 1.563126
1986 | -.2748713 .1605161 -1.71 0.087 -.5896086 .039866
1987 | -.2518984 .1619286 -1.56 0.120 -.5694053 .0656084
1988 | -.2694671 .1632695 -1.65 0.099 -.5896033 .0506691
1989 | -.2437034 .1637741 -1.49 0.137 -.564829 .0774222
1990 | -.2719301 .1631565 -1.67 0.096 -.5918447 .0479844
1991 | -.2678094 .1657425 -1.62 0.106 -.5927946 .0571758
1992 | -.2644613 .1688687 -1.57 0.117 -.5955762 .0666536
1993 | -.2268186 .1775771 -1.28 0.202 -.5750088 .1213717
1994 | -.2336857 .1871741 -1.25 0.212 -.6006937 .1333223
1995 | -.1565192 .1842934 -0.85 0.396 -.5178787 .2048403
1996 | -.0941998 .1831856 -0.51 0.607 -.4533871 .2649876
1997 | -.0987734 .1705863 -0.58 0.563 -.4332562 .2357094
1998 | -.153142 .1651962 -0.93 0.354 -.477056 .170772
1999 | -.1879275 .1673238 -1.12 0.261 -.5160133 .1401583
2000 | -.2191384 .168636 -1.30 0.194 -.5497971 .1115204
2001 | -.2982059 .1663436 -1.79 0.073 -.6243697 .0279579
2002 | -.2929308 .1691806 -1.73 0.083 -.6246573 .0387958
2003 | -.2265498 .1688181 -1.34 0.180 -.5575655 .1044659
2004 | -.1615362 .1757443 -0.92 0.358 -.5061328 .1830603
2005 | -.1664573 .1778879 -0.94 0.349 -.515257 .1823425
2006 | -.2180559 .1818557 -1.20 0.231 -.5746354 .1385237
2007 | -.2708258 .1801942 -1.50 0.133 -.6241476 .0824961
2008 | -.4241269 .1813867 -2.34 0.019 -.7797869 -.0684669
2009 | -.4862781 .1715074 -2.84 0.005 -.822567 -.1499893
2010 | -.4239721 .1710789 -2.48 0.013 -.7594209 -.0885234
2011 | -.4113883 .1736931 -2.37 0.018 -.7519628 -.0708138
2012 | -.3568736 .1839606 -1.94 0.052 -.7175806 .0038333
2013 | -.2236058 .2208678 -1.01 0.311 -.6566798 .2094681
2014 | -.3258896 .2045508 -1.59 0.111 -.7269695 .0751903
2015 | -.3724304 .2121414 -1.76 0.079 -.7883938 .043533
2016 | -.3681389 .2272705 -1.62 0.105 -.8137672 .0774894
2017 | -.4425096 .2373986 -1.86 0.062 -.9079969 .0229777
2018 | -.4737445 .2381267 -1.99 0.047 -.9406595 -.0068296
2019 | -.2071411 .2362073 -0.88 0.381 -.6702925 .2560103
2020 | -.188094 .2496542 -0.75 0.451 -.6776119 .3014238
2021 | -.1728412 .2453747 -0.70 0.481 -.6539679 .3082855
2022 | 0 (omitted)
|
_cons | -15.03153 1.55446 -9.67 0.000 -18.07949 -11.98357
-----------------------------------------------------------------------------------------------------------

C)
. margins, dydx(c.log_gdp_capita) at(c.log_gdp_capita=($MyMin($Grid)15) dum1986=(0))

Average marginal effects Number of obs = 3,088
Model VCE: Robust

Expression: Linear prediction, predict()
dy/dx wrt: log_gdp_capita
1._at: log_gdp_capita = 5.933783
dum1986 = 0
2._at: log_gdp_capita = 6.024445
dum1986 = 0
3._at: log_gdp_capita = 6.115107
dum1986 = 0
4._at: log_gdp_capita = 6.205769
dum1986 = 0
5._at: log_gdp_capita = 6.296431
dum1986 = 0
6._at: log_gdp_capita = 6.387093
dum1986 = 0
7._at: log_gdp_capita = 6.477756
dum1986 = 0
8._at: log_gdp_capita = 6.568418
dum1986 = 0
9._at: log_gdp_capita = 6.65908
dum1986 = 0
10._at: log_gdp_capita = 6.749742
dum1986 = 0
11._at: log_gdp_capita = 6.840404
dum1986 = 0
12._at: log_gdp_capita = 6.931066
dum1986 = 0
13._at: log_gdp_capita = 7.021729
dum1986 = 0
14._at: log_gdp_capita = 7.112391
dum1986 = 0
15._at: log_gdp_capita = 7.203053
dum1986 = 0
16._at: log_gdp_capita = 7.293715
dum1986 = 0
17._at: log_gdp_capita = 7.384377
dum1986 = 0
18._at: log_gdp_capita = 7.47504
dum1986 = 0
19._at: log_gdp_capita = 7.565702
dum1986 = 0
20._at: log_gdp_capita = 7.656364
dum1986 = 0
21._at: log_gdp_capita = 7.747026
dum1986 = 0
22._at: log_gdp_capita = 7.837688
dum1986 = 0
23._at: log_gdp_capita = 7.92835
dum1986 = 0
24._at: log_gdp_capita = 8.019013
dum1986 = 0
25._at: log_gdp_capita = 8.109675
dum1986 = 0
26._at: log_gdp_capita = 8.200337
dum1986 = 0
27._at: log_gdp_capita = 8.290999
dum1986 = 0
28._at: log_gdp_capita = 8.381661
dum1986 = 0
29._at: log_gdp_capita = 8.472323
dum1986 = 0
30._at: log_gdp_capita = 8.562986
dum1986 = 0
31._at: log_gdp_capita = 8.653648
dum1986 = 0
32._at: log_gdp_capita = 8.74431
dum1986 = 0
33._at: log_gdp_capita = 8.834972
dum1986 = 0
34._at: log_gdp_capita = 8.925634
dum1986 = 0
35._at: log_gdp_capita = 9.016297
dum1986 = 0
36._at: log_gdp_capita = 9.106959
dum1986 = 0
37._at: log_gdp_capita = 9.197621
dum1986 = 0
38._at: log_gdp_capita = 9.288283
dum1986 = 0
39._at: log_gdp_capita = 9.378945
dum1986 = 0
40._at: log_gdp_capita = 9.469607
dum1986 = 0
41._at: log_gdp_capita = 9.56027
dum1986 = 0
42._at: log_gdp_capita = 9.650932
dum1986 = 0
43._at: log_gdp_capita = 9.741594
dum1986 = 0
44._at: log_gdp_capita = 9.832256
dum1986 = 0
45._at: log_gdp_capita = 9.922918
dum1986 = 0
46._at: log_gdp_capita = 10.01358
dum1986 = 0
47._at: log_gdp_capita = 10.10424
dum1986 = 0
48._at: log_gdp_capita = 10.1949
dum1986 = 0
49._at: log_gdp_capita = 10.28557
dum1986 = 0
50._at: log_gdp_capita = 10.37623
dum1986 = 0
51._at: log_gdp_capita = 10.46689
dum1986 = 0
52._at: log_gdp_capita = 10.55755
dum1986 = 0
53._at: log_gdp_capita = 10.64822
dum1986 = 0
54._at: log_gdp_capita = 10.73888
dum1986 = 0
55._at: log_gdp_capita = 10.82954
dum1986 = 0
56._at: log_gdp_capita = 10.9202
dum1986 = 0
57._at: log_gdp_capita = 11.01086
dum1986 = 0
58._at: log_gdp_capita = 11.10153
dum1986 = 0
59._at: log_gdp_capita = 11.19219
dum1986 = 0
60._at: log_gdp_capita = 11.28285
dum1986 = 0
61._at: log_gdp_capita = 11.37351
dum1986 = 0
62._at: log_gdp_capita = 11.46418
dum1986 = 0
63._at: log_gdp_capita = 11.55484
dum1986 = 0
64._at: log_gdp_capita = 11.6455
dum1986 = 0
65._at: log_gdp_capita = 11.73616
dum1986 = 0
66._at: log_gdp_capita = 11.82682
dum1986 = 0
67._at: log_gdp_capita = 11.91749
dum1986 = 0
68._at: log_gdp_capita = 12.00815
dum1986 = 0
69._at: log_gdp_capita = 12.09881
dum1986 = 0
70._at: log_gdp_capita = 12.18947
dum1986 = 0
71._at: log_gdp_capita = 12.28013
dum1986 = 0
72._at: log_gdp_capita = 12.3708
dum1986 = 0
73._at: log_gdp_capita = 12.46146
dum1986 = 0
74._at: log_gdp_capita = 12.55212
dum1986 = 0
75._at: log_gdp_capita = 12.64278
dum1986 = 0
76._at: log_gdp_capita = 12.73345
dum1986 = 0
77._at: log_gdp_capita = 12.82411
dum1986 = 0
78._at: log_gdp_capita = 12.91477
dum1986 = 0
79._at: log_gdp_capita = 13.00543
dum1986 = 0
80._at: log_gdp_capita = 13.09609
dum1986 = 0
81._at: log_gdp_capita = 13.18676
dum1986 = 0
82._at: log_gdp_capita = 13.27742
dum1986 = 0
83._at: log_gdp_capita = 13.36808
dum1986 = 0
84._at: log_gdp_capita = 13.45874
dum1986 = 0
85._at: log_gdp_capita = 13.54941
dum1986 = 0
86._at: log_gdp_capita = 13.64007
dum1986 = 0
87._at: log_gdp_capita = 13.73073
dum1986 = 0
88._at: log_gdp_capita = 13.82139
dum1986 = 0
89._at: log_gdp_capita = 13.91205
dum1986 = 0
90._at: log_gdp_capita = 14.00272
dum1986 = 0
91._at: log_gdp_capita = 14.09338
dum1986 = 0
92._at: log_gdp_capita = 14.18404
dum1986 = 0
93._at: log_gdp_capita = 14.2747
dum1986 = 0
94._at: log_gdp_capita = 14.36536
dum1986 = 0
95._at: log_gdp_capita = 14.45603
dum1986 = 0
96._at: log_gdp_capita = 14.54669
dum1986 = 0
97._at: log_gdp_capita = 14.63735
dum1986 = 0
98._at: log_gdp_capita = 14.72801
dum1986 = 0
99._at: log_gdp_capita = 14.81868
dum1986 = 0
100._at: log_gdp_capita = 14.90934
dum1986 = 0
101._at: log_gdp_capita = 15
dum1986 = 0

--------------------------------------------------------------------------------
| Delta-method
| dy/dx std. err. t P>|t| [95% conf. interval]
---------------+----------------------------------------------------------------
log_gdp_capita |
_at |
1 | . (not estimable)
2 | . (not estimable)
3 | . (not estimable)
4 | . (not estimable)
5 | . (not estimable)
6 | . (not estimable)
7 | . (not estimable)
8 | . (not estimable)
9 | . (not estimable)
10 | . (not estimable)
11 | . (not estimable)
12 | . (not estimable)
13 | . (not estimable)
14 | . (not estimable)
15 | . (not estimable)
16 | . (not estimable)
17 | . (not estimable)
18 | . (not estimable)
19 | . (not estimable)
20 | . (not estimable)
21 | . (not estimable)
22 | . (not estimable)
23 | . (not estimable)
24 | . (not estimable)
25 | . (not estimable)
26 | . (not estimable)
27 | . (not estimable)
28 | . (not estimable)
29 | . (not estimable)
30 | . (not estimable)
31 | . (not estimable)
32 | . (not estimable)
33 | . (not estimable)
34 | . (not estimable)
35 | . (not estimable)
36 | . (not estimable)
37 | . (not estimable)
38 | . (not estimable)
39 | . (not estimable)
40 | . (not estimable)
41 | . (not estimable)
42 | . (not estimable)
43 | . (not estimable)
44 | . (not estimable)
45 | . (not estimable)
46 | . (not estimable)
47 | . (not estimable)
48 | . (not estimable)
49 | . (not estimable)
50 | . (not estimable)
51 | . (not estimable)
52 | . (not estimable)
53 | . (not estimable)
54 | . (not estimable)
55 | . (not estimable)
56 | . (not estimable)
57 | . (not estimable)
58 | . (not estimable)
59 | . (not estimable)
60 | . (not estimable)
61 | . (not estimable)
62 | . (not estimable)
63 | . (not estimable)
64 | . (not estimable)
65 | . (not estimable)
66 | . (not estimable)
67 | . (not estimable)
68 | . (not estimable)
69 | . (not estimable)
70 | . (not estimable)
71 | . (not estimable)
72 | . (not estimable)
73 | . (not estimable)
74 | . (not estimable)
75 | . (not estimable)
76 | . (not estimable)
77 | . (not estimable)
78 | . (not estimable)
79 | . (not estimable)
80 | . (not estimable)
81 | . (not estimable)
82 | . (not estimable)
83 | . (not estimable)
84 | . (not estimable)
85 | . (not estimable)
86 | . (not estimable)
87 | . (not estimable)
88 | . (not estimable)
89 | . (not estimable)
90 | . (not estimable)
91 | . (not estimable)
92 | . (not estimable)
93 | . (not estimable)
94 | . (not estimable)
95 | . (not estimable)
96 | . (not estimable)
97 | . (not estimable)
98 | . (not estimable)
99 | . (not estimable)
100 | . (not estimable)
101 | . (not estimable)
--------------------------------------------------------------------------------

[Clyde Schechter]

Thank you. The problem is now clear. Your variable dum_1986 is colinear with the i.year variables, because dum_1986 is an indicator for the year 1950 through 1985. Because of that colinearity, Stata omitted 2022.year from the analysis (observe: "note: 2022.year omitted because of collinearity.")

When you have a model which contains colinear variables, the model is unidentified. In order to identify the model, some constraint(s) must be applied. In this case, Stata chose to do this by omitting 2022.year from the calculation. It could just as well have omitted some other year indicator, or it could have omitted dum_1986. And there are many other possibilities. All of these constraints, however, lead to different regression results for the coefficients of the year variables and dum_1986. In other words, in this kind of model, the coefficients of the year variables and dum_1986 are simply artifacts of how the colinearity among them was removed, and are therefore meaningless. That is why -margins- refuses to take you seriously when you try to use dum_1986 in the -margins- command, because it knows from the regression results that dum_1986 is now a meaningless variable in your model.

Moral of the story: in a two-way fixed effects model, which is what you have here, you must not include any variables that indicate some subset of the years (like dum_1986). If dum_1986 is a key aspect of your research question, the closest alternative is to keep dum_1986 but remove year fixed effects and go with just id level fixed effects. But you cannot have both dum_1986 and the year effects in the model.

[Rasmus Madsen]

Thank you. The problem is now clear. Your variable dum_1986 is colinear with the i.year variables, because dum_1986 is an indicator for the year 1950 through 1985. Because of that colinearity, Stata omitted 2022.year from the analysis (observe: "note: 2022.year omitted because of collinearity.")

When you have a model which contains colinear variables, the model is unidentified. In order to identify the model, some constraint(s) must be applied. In this case, Stata chose to do this by omitting 2022.year from the calculation. It could just as well have omitted some other year indicator, or it could have omitted dum_1986. And there are many other possibilities. All of these constraints, however, lead to different regression results for the coefficients of the year variables and dum_1986. In other words, in this kind of model, the coefficients of the year variables and dum_1986 are simply artifacts of how the colinearity among them was removed, and are therefore meaningless. That is why -margins- refuses to take you seriously when you try to use dum_1986 in the -margins- command, because it knows from the regression results that dum_1986 is now a meaningless variable in your model.

Moral of the story: in a two-way fixed effects model, which is what you have here, you must not include any variables that indicate some subset of the years (like dum_1986). If dum_1986 is a key aspect of your research question, the closest alternative is to keep dum_1986 but remove year fixed effects and go with just id level fixed effects. But you cannot have both dum_1986 and the year effects in the model.

Thank you so much, that makes sense! I'll try to figure it out.

[Rasmus Madsen]

Thank you. The problem is now clear. Your variable dum_1986 is colinear with the i.year variables, because dum_1986 is an indicator for the year 1950 through 1985. Because of that colinearity, Stata omitted 2022.year from the analysis (observe: "note: 2022.year omitted because of collinearity.")

When you have a model which contains colinear variables, the model is unidentified. In order to identify the model, some constraint(s) must be applied. In this case, Stata chose to do this by omitting 2022.year from the calculation. It could just as well have omitted some other year indicator, or it could have omitted dum_1986. And there are many other possibilities. All of these constraints, however, lead to different regression results for the coefficients of the year variables and dum_1986. In other words, in this kind of model, the coefficients of the year variables and dum_1986 are simply artifacts of how the colinearity among them was removed, and are therefore meaningless. That is why -margins- refuses to take you seriously when you try to use dum_1986 in the -margins- command, because it knows from the regression results that dum_1986 is now a meaningless variable in your model.

Moral of the story: in a two-way fixed effects model, which is what you have here, you must not include any variables that indicate some subset of the years (like dum_1986). If dum_1986 is a key aspect of your research question, the closest alternative is to keep dum_1986 but remove year fixed effects and go with just id level fixed effects. But you cannot have both dum_1986 and the year effects in the model.

I got one more question, hope you can help.

The things about calculating the margins is that I want to calculate one version where it only uses the log_GDP and the squared log_GDP (which is when dum1986 is 0) and another version there it takes the log_GDP + the interaction term and the same for the squared model - so I can get the effect of GDP before 1986 and from and after 1986 separated.

Is the moral still the same as before or is there a way around this?

[Clyde Schechter]

The moral is still the same and there is no way around it: it's linear algebra. You can get what you want by omitting the i.year variables from the model.