Ben,
I changed the coding of the by(variable) and now the decomposition is
in line with oaxaca. However, without chaning the coding of this by
variable, my results are completely different. Which ones can I rely
on?
On Nov 23, 2007 4:51 AM, Ben Jann <[email protected]> wrote:
> Probably, this is because the low and high groups are interchanged.
> The threefold decomposition depends on this (have a look at the
> formulas in the helpfile).
>
> Say, your oaxaca command is
>
> . oaxaca est1 est2, tf
>
> Then try
>
> . oaxaca est1 est2, tf asis
>
> If it does not change, type
>
> . oaxaca est2 est2, tf asis
>
> Or, alternatively, change the coding of the by-variable in the oaxaca2 call.
>
> ben
>
>
> On Nov 23, 2007 4:46 AM, Tam Phan <[email protected]> wrote:
> > Hi All:
> >
> > Should Oaxaca and Oaxaca2 provide me w similar decomposition results?
> > I tested with regular regress, albeit the total difference is the
> > same, however,
> > when it comes down to the three fold decomposition, they are not the
> > same and the sign changes. Should this be the case?
> >
> > Tam
> > *
> > * 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/
> >
> *
> * 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/
>
--
Tam D. Phan
Department of Economics
College of Business Administration
University of Central Florida
P.O. Box 161400
Orlando, Florida 32816-1400
Fax: 407-823-3269
*
* 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/
