st: Efficient Sequential Matching

[Jonathan Stoeterau <jonathanstoeterau@gmail.com>]

Dear helpful Statalist,

I have panel data of n individuals and t month (unbalanced). In every
month some individuals move, indicated by a dummy (moving). I am
planning to match moving individuals with non-moving individuals in
their respective month. Two questions in this regard:

1. Before matching, I would like to compare group means of treatment
and control group, as in

quietly estpost ttest variables' by(moving) unequal
esttab ., wide

However, this compares individuals moving in one month with everyone
else in every month (incl. themselves earlier and later). In the best
case I would like to compare means with other individuals in the
respective month only and then aggregate test statistics afterwards.
Is there an efficient way to do this?

2. Matching should follow a similar algorithm – matching only
individuals in the respective month. Is there an easier, faster way
then dropping all individuals month by month with -foreach- and then
merging the matched dataset for each month?

Thank you all very much in advance.

*
*   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/