This example is taken from Stata’s manual on Heckman model. The data set https://stats.idre.ucla.edu/wp-content/uploads/2016/02/womenwk.sas7bdat is in SAS 8 format and can be accessed following the link.
libname in 'd:datasas'; data womenwk; set in.womenwk; sel = (wage~=.); run; proc qlim data = womenwk ; model sel = married children education age /discrete; model wage = education age /select(sel=1); run;
The QLIM Procedure
Summary Statistics of Continuous Responses
N Obs N Obs Standard Lower Upper Lower Upper Variable N Mean Error Type Bound Bound Bound Bound
wage 1343 23.69217 6.305374 Regular
Discrete Response Profile of sel
Index Value Frequency Percent
1 0 657 32.85 2 1 1343 67.15
Model Fit Summary
Number of Endogenous Variables 2 Endogenous Variable sel wage Number of Observations 2000 Log Likelihood -5178 Maximum Absolute Gradient 0.0001161 Number of Iterations 24 AIC 10377 Schwarz Criterion 10433
Algorithm converged.
Parameter Estimates
Standard Approx Parameter Estimate Error t Value Pr > |t|
wage.Intercept 0.485795 1.077038 0.45 0.6520 wage.EDUCATION 0.989953 0.053256 18.59 <.0001 wage.AGE 0.213129 0.020603 10.34 <.0001 _Sigma.wage 6.004795 0.165720 36.23 <.0001 sel.Intercept -2.491015 0.189340 -13.16 <.0001 sel.MARRIED 0.445171 0.067395 6.61 <.0001 sel.CHILDREN 0.438707 0.027783 15.79 <.0001 sel.EDUCATION 0.055732 0.010735 5.19 <.0001 sel.AGE 0.036510 0.004153 8.79 <.0001 _Rho 0.703504 0.051227 13.73 <.0001