Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: difference between interaction created with # and created with xi3


From   uso stata <usostata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: difference between interaction created with # and created with xi3
Date   Mon, 23 Aug 2010 15:49:23 -0400

Hello statlist,
I am currently using a linear model specification and I am fiddling
with interactions.

I have three BINARY regressors a b c

When I use the command a#b#c stata spits out 7 interactions:

a b c
0 0 1
0 1 0
0 1 1
1 0 0
1 0 1
1 1 0
1 1 1

when I use xi3: and a*b*c I get:
a
b
c
_Iaxb
_Iaxc
_Ibxc
_axbxc

While the main effects are identical, i.e. the coefficient of "0 0 1"
is the same as the coefficient of "c", the interaction terms are all
different.

P

The xi3 methods generates the same coefficient as the a##b##c method,
that is equivalent to i.a i.b i.c a#b#c

So my question is what is the specification that stata creates for
a#b#c? How do I interpret for example "0 1 1", I thought it was the
same as the interaction of b and c, but it seems not.
*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index