Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: oaxaca2 command


From   "Ben Jann" <ben.jann@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: oaxaca2 command
Date   Tue, 21 Nov 2006 17:56:37 +0100

The problem is that "_se" is stored in the main tobit equation in
Stata versions prior version 9. This fools oaxaca. oaxaca assumes
"_se" be another covariate in the model and tries to compute its mean.
Note that oaxaca2 will not work even if you use Stata 9 because the
version is reset to 8.2 internally. I will change oaxaca2 so that it
does not reset version. In the meanwhile, estimate and store your
models under version 9 and then apply oaxaca to the stored estimates
(not oaxaca2). In your case:

version 9
tobit lasset_s sratio ... if sexo==0, ll
est sti sexo0
tobit lasset_s sratio ... if sexo==1, ll
est sti sexo1
oaxaca sexo0 sexo1

Hope this helps.
ben

On 11/21/06, Maria Cecilia Calderon <cmaria@sas.upenn.edu> wrote:
Dear Ben and all,
                     I am trying to run an oaxaca decomposition using a Tobit
model for both groups. I obtain the model estimates, however, an error message
appears before the decomposition (variable _se not found). Below is my output.
Do you think you can help me?
Thanks a lot!
Cecilia


oaxaca2 lasset_s sratio cpiyear edumai edupai tierrapas_ss comun_d2 comun_d3
comun_d4, by(sexo) noisily cmd(tobit) cmdopts(ll)
[...]
variable _se not found

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index