st: Interaction terms for panel data

[Simon Pickert <simon.pickert@t-online.de>]

Dear Stata listers,

how do I construct interaction terms in a fixed-effects regression in order to test for differences between certain categories of respondents (e.g. male vs. female)? With respondents I refer to the ones who make up your (panel) data set.

Example question: The relation between X and Y is stronger for males than for females.

I can’t find any literature or information on interaction terms for panel data..maybe because there can’t be interaction terms in panel data to answer this kind of question?
The problem I see here is, that panel data only allows for one observation per time-unit. So it seems impossible to me to construct interaction terms, because you can have observations from both adults and children on a particular day.

Any ideas or references to literature on this is greatly appreciated.

Thank you
Simon Pickert
*
*   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/