--- On Thu, 12/11/09, Andrea Rispoli wrote:
> 1. I learned that since Panel poisson is a non-linear model
> I cannot make considerations on the coefficients and I have
> to calculate marginal effect. Would " mfx compute" be the
> appropriate command?
>
> 2. The marginal effect of the interaction term cannot be
> directly interpreted and I have to draw the graph: is this
> correct? Is there any specific command that I should use?
Not necesarily, it is probably easier to just specify the
-irr- option in -xtpoisson-, this will give you the
incidence rate ratios. This will also allow you to directly
interpret the interaction terms (only now on a multiplicative
scale rather than additive). This also has the advantage of
not having to fix all the explanatory variables and the
subject specific constants to a specific value.
For more see J. Scott Long & Jeremy Freese (2006) Regression
Models for Categorical Dependent Variables Using Stata, 2nd
Edition. College Station, TX: Stata Press.
http://www.stata-press.com/books/regmodcdvs.html
Hope this helps,
Maarten
--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://www.maartenbuis.nl
--------------------------
*
* 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/
