Are you sure it didn't drop a dummy variable?
In -reg- if you have full set of dummies and a constant it will report the
dropped dummy variable in the table of estimated coefficients, something
like:
webuse grunfeld,clear
tab time, gen(D)
reg inve mval D*
<snip>
mvalue | .1398271 .0059889 23.35 0.000 .1280092 .151645
D1 | (dropped)
D2 | -23.16908 49.15807 -0.47 0.638 -120.173 73.8348
<snip>
However, in -xtreg- it will report the dropped variable at the top of the
output:
. xtreg inve mval D*, re
note: D5 dropped due to collinearity
<snip>
Scott
> -----Original Message-----
> From: [email protected] [mailto:owner-
> [email protected]] On Behalf Of Cordula Stolberg
> Sent: Tuesday, June 07, 2005 8:26 AM
> To: [email protected]
> Subject: st: Dummy Variable Trap
>
> Hello All,
>
> I have a panel regression, which I first run as a random effects
> regression
> and then as pooled OLS. I have yearly observations and add a time dummy
> for
> each year (the time dummies are also used for an interaction term with
> another independent variable). I know that to avoid the dummy variable
> trap,
> I should drop one of the dummies. I also know that Stata otherwise drops
> one
> of the time dummies. Now I have estimated the random effects regression
> and
> the pooled OLS regression and kept (by mistake) all of the time dummies.
> However, instead of dropping one of the dummies, Stata kept all and
> delivered results. What am I missing here?
>
> Thanks,
> Cordula
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
