st: Fama/MacBeth approach for mediation models

["Florian Christian Esser" <florian.esser@tu-dortmund.de>]

Hi everyone,

I want to use panel data for a moderated mediation analysis and since I do
not just want to calculate mediation models cross-sectionally, I was
trying to implement Fama/MacBeth standard errors for my mediation model.
Does anyone know how to implement this approach?
I am using the .ado files from Peterson and Hoechle
(http://www.kellogg.northwestern.edu/faculty/petersen/htm/papers/se/se_programming.htm).
Whereas I can of course build the "regular" regressions as in, e.g., Hayes
(2013), I don't know how I would approach quantifying and testing the
direct, indirect and total effect. And I am not even talking about
bootstrapping...

Any help/suggestions are greatly appreciated.

*
*   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/