Re: st: Margins for interactions terms

[Nick Cox <njcoxstata@gmail.com>]

You asked a similar question before.

http://www.stata.com/statalist/archive/2013-12/msg00300.html

-reg2hdfe- is a user-written program from SSC, as you are asked to
explain. Evidently the author is not an active member of Statalist;
many users, naturally, are not.

Adding code to return e(sample) is documented at -help ereturn-. You
would need to clone the program, find the appropriate place to add
code and take responsibility for what happens. Alternatively, try
emailing the author directly.

Nick
njcoxstata@gmail.com


On 9 December 2013 16:45, emanuele mazzini <madsoenistata@gmail.com> wrote:
> Dear all,
>
> I am estimating a gravity model with exporter-year and importer-year
> FE with an interaction term  and the reg2hdfe command, as follows:
>
>
> reg2hdfe ln(trade_flow) ln(gdp_o) ln(gdp_d) ln(distance) x1 x2 x1*x2,
> id1(it) id2(it)
>
> where it and jt are defined as:
>
> egen it = group (iso_o year)
> egen jt = group (iso_d year)
>
> After I obtain the estimation results, I try to run the margins
> command as to obtain the marginal effect of x1 on the dependent
> variable lx and draw a marginsplot.
>
> Unfortunately, when I run -margins- Stata returns me the following error:
>
> e(sample) does not identify the estimation sample.
>
> which seems a bit strange to me, because -ereturn list- gives me the e(N)
> function.
>
> Anyone knows how to add the e(sample) to the list of functions that
> -reg2hdfe- returns?
>
> Thank you in advance,
>
> Regards,
>
> Emanuele.
> *
> *   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/
*
*   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/