Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Structural breaks in panel data
From
Maarten Buis <[email protected]>
To
[email protected]
Subject
Re: st: Structural breaks in panel data
Date
Wed, 4 Jul 2012 20:46:30 +0200
On Wed, Jul 4, 2012 at 2:38 PM, Tanja Berg <[email protected]> wrote:
> I implemented four different Dummy variables (P2-P5) for five different periods.
>
> So my regression looks like the following:
> xtreg depvar x1 x2 P2 P3 P4 P5 x1*P2 x1*P3 x1*P4 x1*P5 x2*P2 x2*P3 x2*P4 x2*P5, fe
>
> x1 and x2 are my explanatory variables.
> If the coefficient of the interaction variable (e.g. x1*P2) is significant I have an effect here in this period.
That is incorrect. It means that you cannot reject the hypothesis the
effect of x1 in period 1 is the same as the effect of x1 in period 2.
> But what do the coefficients of the Dummy variables itself tell me? What if they are insignificant?
It gives you the difference in effect between the reference period
(period 1) and the period represented by the interaction effects.
> If I only want to look at one explanatory variable, do I have to include the Dummy variables here as well or is it sufficient to look only at the interaction variables like following:
> xtreg depvar x1 x1*P2 x1*P3 x1*P4 x1*P5, fe
you need to include all main effects, i.e. you also need to include
the period dummies.
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/