Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

st: ivreg2: interacting the endogenous regressor

From   Jana von Stein <>
Subject   st: ivreg2: interacting the endogenous regressor
Date   Wed, 23 May 2012 14:17:14 -0400

Hi -- thanks for the idea, Kit (Go Blue!). Unfortunately, even decent instruments are very hard to find. Essentially, I am trying to test whether the effect of the instrumented variable depends on the values that the variable x1 takes on. Is there some way to do this that doesn't involve trying to find a second instrument (which is near- impossible)? I've split x1 into "low" and "high" and run separate regressions, but that doesn't allow cross-estimation comparisons, right?

many thanks,
On May 23, 2012, at 5:52 AM, Christopher Baum wrote:

On May 23, 2012, at 8:33 AM, Jana wrote:

I need to interact my endogenous variable (for which I'm
instrumenting) with one of the exogenous variables in my step2
equation. Is there a straightforward way to do this using ivreg2? I'm
reluctant to do it by hand because I'm not certain whether the
standard errors would be right; I am also a bit of a novice on iv
regression and so would like to be able to take advantage of ivreg2's
already-programmed post-estimation tests, etc.

You need an additional instrument to do that: if you have y2 and x1*y2 in the regression (y endog, x exog) you need to form the instruiments (z1 z1*x1), where z1 is the excluded instrument. If z1 is an appropriate instrument, and x1 is truly exogenous, then the product of z1*x1 will also be

Go Blue!

Kit Baum
co-author ivreg2

Kit Baum   |   Boston College Economics & DIW Berlin   |
An Introduction to Stata Programming |
 An Introduction to Modern Econometrics Using Stata  |

*   For searches and help try:

*   For searches and help try:

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