Bookmark and Share

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

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: mfx censored ordered probit

From   Marina Mastrorillo <[email protected]>
To   <[email protected]>
Subject   st: mfx censored ordered probit
Date   Sat, 27 Aug 2011 19:51:04 +0100

I'm Marina. I used the do.file written by Edoardo (found online) to
estimate the impact of migration on education using a censored ordered
probit. Many thanks to Edoardo! 
Now I should estimate the marginal effects. Could please anyone help me,
or give me any suggestion? Thanks a lot. 
Best regards,
PS: Below there are the exact commands I used.
/*define school attainment dependent variable*/
gen school0 =schooling==0 
gen school1 =schooling==1 
gen school2 =schooling==2 

gen school3 =schooling==3
gen school4 =schooling==4
gen school5 =schooling==5
gen unc=enrol==0 
//define uncensored observation: children not attending school

program define maxcens
args lnf1 theta1 cut1 cut2 cut3 cut4 cut5 
qui replace `lnf1' = unc* (ln(($ML_y1*normprob(`cut1'-`theta1'))+($ML_y2*
(1-normprob(`cut5' -`theta1')))))+ enrol* (ln(($ML_y1)+($ML_y2*
ml model lf maxcens (school0 school1 school2 school3 school4 school5 =
migcop sex age_head agesq_head num_hh sex_head years_05 years_69 years_1013
years_1417 years_1822 male_mig_hh female_mig_hh school_mighh hh_greece
hh_italy sch_head costal central mountain soc_cap dist_ind m07_q05 reshat,
nocons) /cut1 /cut2 /cut3 /cut4 /cut5 
ml search, repeat(100)
ml maximize, difficult

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index