Bookmark and Share

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]

st: xtreg postestimation "test" or "lincom" using fatorial variables


From   Jason Park <[email protected]>
To   [email protected]
Subject   st: xtreg postestimation "test" or "lincom" using fatorial variables
Date   Wed, 21 Aug 2013 23:41:59 +0800

Hi,

I have D as the dependent variable, and three independent variables A,
B, and C where B and C are indicator (dummy) variables.
I ran a fixed-effects regression: "xtreg D A c.A#i.B c.A#i.C
c.A#i.B#i.C, fe vce(robust)" in order to test the effect of A on D,
particularly the effect conditional on B and C.
However, the biggest challenge I have here is that I don't know how to
test the marginal significance between c.A#i.B#i.C and c.A#i.B.
The coefficient on c.A#i.B#i.C itself indicates the conditional effect
jointly on B = 1 & C = 1 as opposed to B = 0 & C = 0, but what I'm
interested in is the cases between B = 1 & C = 1 and B = 1 & C = 0.
This is the closest I could manage to reach: "lincom c.A#1.B#1.C -
c.A#1.B#0.C" or "test c.A#1.B#1.C - c.A#1.B#0.C = 0"
But the result is just the same as testing the single coefficient on
c.A#i.B#i.C in the above regression model.
Can anyone please advise how to test this in the given setting?

Thanks in advance, J
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index