Examples from the help file
The examples at the end of help meinequality, run in full
The examples from the end of the help file, in order, on Stata’s nlsw88 data.
Single model
sysuse nlsw88, clear(NLSW, 1988 extract)
. reg wage i.race c.age i.married Source | SS df MS Number of obs = 2,246
-------------+---------------------------------- F(4, 2241) = 8.50
Model | 1111.30875 4 277.827187 Prob > F = 0.0000
Residual | 73256.6587 2,241 32.689272 R-squared = 0.0149
-------------+---------------------------------- Adj R-squared = 0.0132
Total | 74367.9674 2,245 33.1260434 Root MSE = 5.7175
------------------------------------------------------------------------------
wage | Coefficient Std. err. t P>|t| [95% conf. interval]
-------------+----------------------------------------------------------------
race |
Black | -1.459 0.283 -5.157 0.000 -2.014 -0.904
Other | 0.463 1.130 0.409 0.682 -1.754 2.679
|
age | -0.084 0.040 -2.122 0.034 -0.161 -0.006
|
married |
Married | -0.782 0.258 -3.035 0.002 -1.287 -0.277
_cons | 11.928 1.575 7.572 0.000 8.839 15.017
------------------------------------------------------------------------------
. meinequality raceME Inequality Estimates (N = 2246)
| Estimate Std. err. z P>|z|
-------------------------+-----------------------------------------------
ME Inequality |
race | 1.153 0.588 1.961 0.050 meinequality race, unweightedME Inequality Estimates (N = 2246)
| Estimate Std. err. z P>|z|
-------------------------+-----------------------------------------------
Unwgt. ME Inequality |
race | 1.281 0.765 1.675 0.094 Compare across two models on the same sample
logit union i.race, vce(robust)Iteration 0: Log pseudolikelihood = -1046.6242
Iteration 1: Log pseudolikelihood = -1040.2968
Iteration 2: Log pseudolikelihood = -1040.2692
Iteration 3: Log pseudolikelihood = -1040.2692
Logistic regression Number of obs = 1,878
Wald chi2(2) = 12.97
Prob > chi2 = 0.0015
Log pseudolikelihood = -1040.2692 Pseudo R2 = 0.0061
------------------------------------------------------------------------------
| Robust
union | Coefficient std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
race |
Black | 0.406 0.117 3.466 0.001 0.177 0.636
Other | 0.554 0.438 1.265 0.206 -0.305 1.412
|
_cons | -1.247 0.065 -19.096 0.000 -1.375 -1.119
------------------------------------------------------------------------------est store basemod
logit union i.race c.age i.married, vce(robust)Iteration 0: Log pseudolikelihood = -1046.6242
Iteration 1: Log pseudolikelihood = -1038.8701
Iteration 2: Log pseudolikelihood = -1038.8354
Iteration 3: Log pseudolikelihood = -1038.8354
Logistic regression Number of obs = 1,878
Wald chi2(4) = 15.82
Prob > chi2 = 0.0033
Log pseudolikelihood = -1038.8354 Pseudo R2 = 0.0074
------------------------------------------------------------------------------
| Robust
union | Coefficient std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
race |
Black | 0.372 0.120 3.093 0.002 0.136 0.608
Other | 0.553 0.439 1.260 0.208 -0.307 1.412
|
age | 0.011 0.018 0.637 0.524 -0.024 0.046
|
married |
Married | -0.176 0.114 -1.549 0.121 -0.399 0.047
_cons | -1.570 0.711 -2.207 0.027 -2.964 -0.176
------------------------------------------------------------------------------est store medmod. meinequality race, models(basemod medmod)ME Inequality Estimates (N_basemod = 1878 , N_medmod = 1878)
| Estimate Std. err. z P>|z|
-------------------------+-----------------------------------------------
race ME Ineq. |
Model 1 (basemod) | 0.083 0.050 1.661 0.097
Model 2 (medmod) | 0.081 0.050 1.608 0.108
Cross-Model Diff. | 0.002 0.003 0.959 0.337 ME inequality by group
In the second model, with the Diff. row testing whether it differs between the groups:
meinequality race, models(medmod) by(married)ME Inequality Estimates (N = 1878)
| Estimate Std. err. z P>|z|
-------------------------+-----------------------------------------------
ME Inequality |
race(Single) | 0.085 0.052 1.629 0.103
race(Married) | 0.079 0.049 1.593 0.111
race Diff. | 0.006 0.005 1.304 0.192
Diff. = Single - MarriedCompare across distinct samples/groups for two models
logit union i.race c.age if married == 0, vce(robust)Iteration 0: Log pseudolikelihood = -386.41898
Iteration 1: Log pseudolikelihood = -384.38689
Iteration 2: Log pseudolikelihood = -384.38343
Iteration 3: Log pseudolikelihood = -384.38343
Logistic regression Number of obs = 656
Wald chi2(3) = 4.12
Prob > chi2 = 0.2493
Log pseudolikelihood = -384.38343 Pseudo R2 = 0.0053
------------------------------------------------------------------------------
| Robust
union | Coefficient std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
race |
Black | 0.359 0.178 2.016 0.044 0.010 0.709
Other | 0.206 0.845 0.243 0.808 -1.451 1.863
|
age | 0.010 0.029 0.346 0.729 -0.047 0.067
_cons | -1.513 1.161 -1.303 0.193 -3.788 0.763
------------------------------------------------------------------------------est store notmar
logit union i.race c.age if married == 1, vce(robust)Iteration 0: Log pseudolikelihood = -657.71228
Iteration 1: Log pseudolikelihood = -654.35596
Iteration 2: Log pseudolikelihood = -654.3266
Iteration 3: Log pseudolikelihood = -654.32659
Logistic regression Number of obs = 1,222
Wald chi2(3) = 6.95
Prob > chi2 = 0.0734
Log pseudolikelihood = -654.32659 Pseudo R2 = 0.0051
------------------------------------------------------------------------------
| Robust
union | Coefficient std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
race |
Black | 0.380 0.163 2.330 0.020 0.060 0.699
Other | 0.692 0.512 1.351 0.177 -0.312 1.695
|
age | 0.012 0.022 0.518 0.605 -0.032 0.056
_cons | -1.762 0.887 -1.986 0.047 -3.500 -0.023
------------------------------------------------------------------------------est store marry. meinequality race, models(notmar marry) groupsME Inequality Estimates (N_notmar = 656 , N_marry = 1222)
| Estimate Std. err. z P>|z|
-------------------------+-----------------------------------------------
race ME Ineq. |
Model 1 (notmar) | 0.055 0.032 1.706 0.088
Model 2 (marry) | 0.097 0.061 1.603 0.109
Cross-Model Diff. | -0.042 0.069 -0.619 0.536 Nominal or ordinal outcome models
One ME inequality is reported per outcome category.
mlogit industry i.race c.ageIteration 0: Log likelihood = -4225.5484
Iteration 1: Log likelihood = -4179.1623
Iteration 2: Log likelihood = -4175.8438
Iteration 3: Log likelihood = -4175.3653
Iteration 4: Log likelihood = -4175.2895
Iteration 5: Log likelihood = -4175.271
Iteration 6: Log likelihood = -4175.2671
Iteration 7: Log likelihood = -4175.2663
Iteration 8: Log likelihood = -4175.2661
Iteration 9: Log likelihood = -4175.2661
Iteration 10: Log likelihood = -4175.2661
Multinomial logistic regression Number of obs = 2,232
LR chi2(33) = 100.56
Prob > chi2 = 0.0000
Log likelihood = -4175.2661 Pseudo R2 = 0.0119
-----------------------------------------------------------------------------------------
industry | Coefficient Std. err. z P>|z| [95% conf. interval]
------------------------+----------------------------------------------------------------
Ag_Forestry_Fisheries |
race |
Black | -0.001 0.579 -0.002 0.999 -1.136 1.134
Other | -15.006 3747.262 -0.004 0.997 -7359.505 7329.494
|
age | 0.074 0.080 0.931 0.352 -0.082 0.231
_cons | -6.809 3.207 -2.123 0.034 -13.096 -0.523
------------------------+----------------------------------------------------------------
Mining |
race |
Black | -14.797 1374.780 -0.011 0.991 -2709.318 2679.723
Other | -15.095 7082.897 -0.002 0.998 -1.39e+04 13867.127
|
age | -0.257 0.198 -1.297 0.195 -0.645 0.131
_cons | 4.785 7.398 0.647 0.518 -9.714 19.284
------------------------+----------------------------------------------------------------
Construction |
race |
Black | -0.686 0.547 -1.254 0.210 -1.758 0.386
Other | 0.853 1.066 0.800 0.424 -1.236 2.942
|
age | -0.073 0.063 -1.150 0.250 -0.197 0.051
_cons | -0.402 2.461 -0.163 0.870 -5.226 4.423
------------------------+----------------------------------------------------------------
Manufacturing |
race |
Black | 0.594 0.137 4.353 0.000 0.327 0.862
Other | -0.020 0.589 -0.034 0.973 -1.175 1.134
|
age | -0.020 0.021 -0.980 0.327 -0.061 0.020
_cons | -0.191 0.818 -0.234 0.815 -1.794 1.412
------------------------+----------------------------------------------------------------
Transport_Comm_Utility |
race |
Black | 0.312 0.245 1.273 0.203 -0.168 0.793
Other | -0.102 1.053 -0.097 0.922 -2.166 1.961
|
age | 0.007 0.036 0.203 0.839 -0.064 0.079
_cons | -2.587 1.437 -1.800 0.072 -5.403 0.230
------------------------+----------------------------------------------------------------
Wholesale_Retail_trade |
race |
Black | -0.256 0.160 -1.597 0.110 -0.570 0.058
Other | -15.084 863.578 -0.017 0.986 -1707.666 1677.497
|
age | 0.003 0.021 0.152 0.879 -0.038 0.045
_cons | -0.963 0.839 -1.148 0.251 -2.609 0.682
------------------------+----------------------------------------------------------------
Finance_Ins_Real_estate |
race |
Black | -0.771 0.230 -3.349 0.001 -1.222 -0.320
Other | -0.384 0.774 -0.495 0.620 -1.901 1.134
|
age | -0.050 0.027 -1.883 0.060 -0.102 0.002
_cons | 0.642 1.040 0.617 0.537 -1.397 2.680
------------------------+----------------------------------------------------------------
Business_Repair_svc |
race |
Black | -0.049 0.270 -0.184 0.854 -0.578 0.479
Other | 0.564 0.781 0.722 0.470 -0.967 2.094
|
age | -0.056 0.038 -1.470 0.141 -0.130 0.018
_cons | -0.093 1.475 -0.063 0.950 -2.983 2.798
------------------------+----------------------------------------------------------------
Personal_services |
race |
Black | 1.021 0.221 4.618 0.000 0.587 1.454
Other | 0.092 1.055 0.088 0.930 -1.975 2.159
|
age | 0.012 0.035 0.332 0.740 -0.058 0.081
_cons | -2.954 1.403 -2.105 0.035 -5.705 -0.204
------------------------+----------------------------------------------------------------
Entertainment_Rec_svc |
race |
Black | -0.355 0.643 -0.552 0.581 -1.615 0.905
Other | -15.066 3718.055 -0.004 0.997 -7302.320 7272.188
|
age | 0.090 0.080 1.123 0.262 -0.067 0.247
_cons | -7.358 3.227 -2.280 0.023 -13.683 -1.034
------------------------+----------------------------------------------------------------
Professional_services | (base outcome)
------------------------+----------------------------------------------------------------
Public_administration |
race |
Black | 0.356 0.184 1.933 0.053 -0.005 0.717
Other | 0.641 0.593 1.082 0.279 -0.520 1.803
|
age | -0.005 0.027 -0.187 0.851 -0.058 0.048
_cons | -1.452 1.076 -1.350 0.177 -3.560 0.657
-----------------------------------------------------------------------------------------
. meinequality raceME Inequality Estimates (N = 2232)
| Estimate Std. err. z P>|z|
-------------------------+-----------------------------------------------
race ME Ineq. |
Pr(Ag/Forestry/Fisherie) | 0.004 0.002 1.880 0.060
Pr(Mining) | 0.002 0.001 2.003 0.045
Pr(Construction) | 0.017 0.019 0.890 0.374
Pr(Manufacturing) | 0.059 0.021 2.788 0.005
Pr(Transport/Comm/Utili) | 0.006 0.011 0.503 0.615
Pr(Wholesale/Retail tra) | 0.101 0.009 10.871 0.000
Pr(Finance/Ins/Real est) | 0.043 0.015 2.897 0.004
Pr(Business/Repair svc) | 0.023 0.027 0.839 0.402
Pr(Personal services) | 0.031 0.012 2.623 0.009
Pr(Entertainment/Rec sv) | 0.005 0.002 2.410 0.016
Pr(Professional service) | 0.046 0.051 0.903 0.367
Pr(Public administratio) | 0.049 0.036 1.331 0.183 Bootstrap example
Wrap the model and the command in an r-class program and hand it to bootstrap. (Shown, not run.)
capture program drop boot_mei
program define boot_mei, rclass
reg wage i.race c.age i.married
meinequality race
return scalar w_mei = r(wem11)
end
bootstrap w_mei=r(w_mei), reps(1000): boot_mei