st: Using the results generated by the Matching Programs

[Joseph Kwan <JXK180@bham.ac.uk>]

Dear Statalists,

I used commands Nearest Neighbour Matching ( attnd & attnw ) and radius matching (attr) , Kernel matching method (attk.ado), Stratification method (atts.ado), and they were extremely slow. When I finally ended up with some results, I just could not use the matched observations and generate a control and treatment dummy variable.

I found out from Stata Journal that they would be stored as scalars, and the problem for me was what do I need to do to assess these results.

Scalars
r(nt*) number of treated used in the computation of att*
r(nc*) number of controls used in the computation of att*
r(att*) ATT obtained by att*
r(seatt*) analytical standard error for att*
r(tsatt*) analytical t statistic for att*
r(bseatt*) bootstrapped standard error for att*
r(btsatt*) bootstrapped t statistic for att*
r(mean1) mean outcome of matched treated for attk
r(mean0) mean outcome of matched controls for attk

This is the result from the matching. I would like to use the matched controls and treatment observations and generate a dummy.

n.  treat. n. contr. ATT      Std. Err. t
185 1157   1537.943  1016.874 1.512

These are the commands I am referring to and they are all described in the Stata Journal on the link above.

http://www.stata-journal.com/sjpdf.html?articlenum=st0026

Please help.


*
*   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/