st: nnmatch with bias adjustment

[Erasmo Giambona <e.giambona@gmail.com>]

Dear Statalisters,

In using nnmatch for an application in finance, I am finding out that
the bias adjustment makes quite a big difference in the effect of
interest. In particular, the average treatmetn effect for the treated
(ATT) almost doubles in size after using the bias adjustment. However,
I am having a hard time understanding how to use the biased adjusted
result. In particular, I would like to understand whether the increase
in size of the ATT is due to an increase in the effect for the treated
or to a decrease in size in the effect for the control.

Is there a way to isolate these two effects?

Thanks,

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