Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

st: Oxaca decomposition with Biprobit

 From Fabrizio Colella - 1348010 To statalist@hsphsun2.harvard.edu Subject st: Oxaca decomposition with Biprobit Date Mon, 9 Dec 2013 19:15:09 +0100 (CET)

Dear all,

I am performing an Oaxaca Blinder wage decomposition among women working part-time and women working full-time,
with a double switching model as selection-correction.

To do that I run a  –biprobit- regression to correct for two types of selection:
1) participation vs. non-participation in the labor market
2) full-time vs. part-time – (CONDITIONAL ON PARTICIPATION)
and I insert the two mills ratios as covariates in the –oaxaca- as explained in the the paper: Axel Heitmueller, 2004,
"Public-Private Sector Wage Differentials in Scotland: An Endogenous Switching Model", IZA DP No. 992.

However, my results seem to be not so good, and I am worried if I am doing something wrong in the mills ratios construction.
Can you please check the follow code?

**************************************
*** Selection
biprobit lfp part_time age_25_35 … … health_prob, robust iterate(10)
predict Xb1 if e(sample), xb1
predict Xb2 if e(sample), xb2
gen rho=0.9906379
gen dens1 = normalden(Xb1)
gen dens2 = normalden(Xb2)
gen bin = binormal(Xb2,Xb1,rho)
gen bin_neg = binormal(-Xb2,Xb1,-rho)
gen mills1 = .
gen mills2 = .
replace mills1 = dens1*(normal((Xb2-(rho*Xb1))/((1-(rho^2))^(1/2))))*(bin^(-1)) if part_time_index==1 //part
replace mills2 = dens2*(normal((Xb1-(rho*Xb2))/((1-(rho^2))^(1/2))))*(bin^(-1)) if part_time_index==1 //part
replace mills1 = dens1*(normal(-((Xb2-(rho*Xb1))/((1-(rho^2))^(1/2)))))*(bin_neg^(-1)) if part_time_index==2 //full
replace mills2 = (-dens2)*(normal((Xb1-(rho*Xb2))/((1-(rho^2))^(1/2))))*(bin_neg^(-1)) if  part_time_index==2 //full
*** OB decomposition ***
oaxaca ln_hourly_wage ed_primary … … worker mills1 mills2, by(part_full_only) noisily weight(0)
*******************************

Every suggestion will be highly appreciated.

Thanks
Fabrizio

--------------------------------
Fabrizio Colella
Bocconi University and
Fondazione Rodolfo DeBenedetti
Milan, Italy
fabrizio.colella@studbocconi.it

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/