Hi Rich and John,
thank you very much for your replies.
Usually, I would also leave the main effects in the regression. However, I want to try an approach to control for multilateral resistance in a gravity model with panel data which was suggested by Baldwin and Taglioni (2006). These authors suggest including time varying importer and exporter dummies. They do not include the main effects for this purpose.
I am not sure if you or anyone else on the list is familiar with estimating the gravity model with panel data. There is some literature about the appropriate estimation approach; however, there does not seem to be a consensus in this regard. So I would obviously appreciate any comment in this matter.
Best regards
Phil
Reference:
Baldwin, R.; Taglioni, D. (2006): Gravity for dummies and dummies for gravity equations. NBER Working Paper 12516.
-------- Original-Nachricht --------
> Datum: Wed, 03 Jun 2009 16:45:30 +0200
> Von: John Antonakis <[email protected]>
> An: [email protected]
> Betreff: Re: st: Creating interaction terms between country and time fixed effects
> Hi:
>
> Not including the main effects would be a fatal flaw; the interaction
> terms mean nothing without the main effects. Given the scale dependency
> of the multiplicative terms
> it is essential to first partial out the main effects from the
> multiplicative term in the regression equation.
>
> HTH,
> J.
>
> ____________________________________________________
>
> Prof. John Antonakis
> Associate Dean Faculty of Business and Economics
> University of Lausanne
> Internef #618
> CH-1015 Lausanne-Dorigny
> Switzerland
>
> Tel ++41 (0)21 692-3438
> Fax ++41 (0)21 692-3305
>
> Faculty page:
> http://www.hec.unil.ch/people/jantonakis&cl=en
>
> Personal page:
> http://www.hec.unil.ch/jantonakis
> ____________________________________________________
>
>
>
> On 03.06.2009 15:31, [email protected] wrote:
> > Hello,
> >
> > I try to create interaction terms between exporter and importer fixed
> effects and time fixed effects. My first thought was to use the xi command
> (xi: reg ..... i.ci*i.year i.cj*i.year, where ci and cj are exporters and
> importers). STATA does create the interaction terms, however, the main effects
> always remain in the regression.
> >
> > My question is whether there is an alternative command which I can use
> to create the interaction terms and which allows me to estimate the model
> only with the interactions included and not with the main effects?
> >
> > I would appreciate any help,
> >
> > best regards
> >
> > Phil
> >
> >
> >
> *
> * 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/
--
GMX FreeDSL mit DSL 6.000 Flatrate und Telefonanschluss nur 17,95 Euro/mtl.!
http://dslspecial.gmx.de/freedsl-aktionspreis/?ac=OM.AD.PD003K11308T4569a
*
* 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/
