st: Simple slopes analysis in fixed effects panel regression

["Maartje Groenewegen" <maartjegroenewegen@live.nl>]

Dear list members,

Currently I am conducting a fixed effects panel regression which includes a marginal significant 
(p=0.067) two-way interaction term consisting of two mean centered continuous variables.
I would like to perform a simple slopes analysis on this interaction, but I don’t know how to do it.
I’ve used sslope from Jeffrey Simons, but that does not take the panel character of the data into 
account.
Additionally, I used ‘margins’, but I’m not sure how to interpret the results since the ‘Delta-method 
table’ shows the same values for all statistics for all 12 categories (all are non significant). The 
command I used is this:

‘margins, dydx(IV) at(MODERATOR=(minIV(0.25)maxIV)) vsquish’           (minIV = –1.20, maxIV=1.60)

Thereafter I used the following command, and now the Delta-method table shows all significant results 
for the 24 categories. However, I don’t know how to interpret these results.

‘margins, at(IV=(minIV maxIV) MODERATOR=(minIV(0.25)maxIV)) vsquish’

Could someone assist me with finding the right command to perform a simple slope analysis for fixed 
effects panel regression and how should these results be interpreted?

Thanks a lot!

Maartje 

*
*   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/