If some non-standard application of -xi- is desired, use the first syntax
mentioned in -h xi- and kill (-drop-) the dummies you do not want. Then
insert the ones you do want in your estimation...
HTH
Martin
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Galina An
Sent: Saturday, September 20, 2008 8:52 PM
To: [email protected]
Subject: st: help with: i.region*i.time
Dear stata users,
I was wondering if anybody could help me with the following (seemingly
simple!) problem.
I need to keep interaction dummies in the regression, but not the
dummies themselves.
For example, in the regression:
xi: reg y x i.region*t2 i.region*t3 i.region*t4
I need the interactions between region and time dummies, but not the
region dummies themselves.
Thanks!
--
Galina An
Assistant Professor
Economics Department
Kenyon College
*
* 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/
*
* 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/
