[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: re: missing dummy variable

From   "Wallace, John" <>
To   <>
Subject   st: RE: re: missing dummy variable
Date   Wed, 10 Oct 2007 14:35:05 -0700

Thanks Kit, but the coefficient is the easy part!  It's the confidence
intervals I'm scratching my head over.  Is this something that can be
dug out of the e(V) matrix or the like?

-----Original Message-----
[] On Behalf Of Kit Baum
Sent: Wednesday, October 10, 2007 2:11 PM
Subject: st: re: missing dummy variable

John said

Agreed - although I have occasions in the data I work with where the
grand mean itself is of interest, and the variation of the coefficients
from that mean is useful (hence my interest in -xi3- ).  The problem is
the missing coefficient for one of the indicators.

No problem.  As I said before, the algebra isn't hard--it's just a  
pain to do with -lincom-. Recall that if you have a constant term  
that estimates the grand mean, the sum of dummy coefficients around  
that mean is zero by construction. Therefore the missing coefficient  
is minus the sum of those which you estimate. That is easily seen in  
your first example using e.foreign; the missing coefficient is just  
-1 * the included coefficient.  But as I say doing that for two dozen  
included coeffs. is painful.


Kit Baum, Boston College Economics and DIW Berlin
An Introduction to Modern Econometrics Using Stata:

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index