Using Oaxaca-Blinder analysis I would like to investigate a wage premium for married men. However, I had some problems with STATA command syntax :(
I know there exists "nldecompose" command, but I'm completely lost with varname, omega and other components it includes...
Could you please help me to deal with this?
Thanks in advance!
Hello Dear, varname is the dummy var by( marry status= 1, others = 0) and omega ll(income,age,..) to run this command you can use lineal and non lineal like regress.probit, logit ..
lmincome( logaritmo of income)
ex:
nldecompose, by(marry status) regoutput: regress lnincome educ age
regregoutput( display the 2 ecuacions)
Dont forget that dummy equation is in fact 2 equation in 1.
I hope i help you . good luck !!
Could anyone show an example of nldecompose with multiply imputed data and weights (svy)?
Looking for the command I read that "calculation of sample means are now adjusted for the use of weights and svy commands" . However, I cannot make it work on my imputed & survey data.
I am not sure where to add the mi estimate..I tried the following, but it does not work
mi estimate, cmdok: nldecompose, by(firstg): svy: poisson salary Medium_c Low_c Unrelated_c Misc
Thank you!
Hello, Margarita. You could use directly the Oaxaca command in Stata. I have been using in some works with good results. You can check http://www.stata-journal.com/article.html?article=st0151
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software