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

st: test after iv

From   Daniel Borowczyk Martins <>
Subject   st: test after iv
Date   Mon, 14 Dec 2009 13:25:58 +0000

Dear all,

I have a quick question I could not see answered neither in FAQ, nor
in the statalist posts nor stata help. Thus, it would very helpful if
you could answer me.

I estimate my model with a simple 2sls cross section regression. There
are two groups of observations.

ivregress 2sls prod  $x1list g2 $x2list (ent01 ent012  g2_ent01
g2_ent012 = $z1list $z2list $z3list), vce(robust)

x1list is the list of independent variables

g2  is a dummy for group 2.

x2list includes interactions between all independent variables and the g2 dummy.

I instrument ent01, its quadratic term (ent012) and respective
interactions with g2 (g2_ent01 g2_ent012) with iv counterparts (i.e.
z1list, z2list and z3list).

Now, I want to test overall significance of the interaction terms. How
should I do that with command test?

Will this:

test $x2list g2_ent01 g2_ent012

do the job?

I mean, will the instruments used in estimating the model be also used
in testing overall significance?

Many thanks.
*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index