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: Time invariant variable instrumenting it and intreg

From   "dsrhg dgr" <>
Subject   st: Time invariant variable instrumenting it and intreg
Date   Thu, 2 May 2013 02:11:13 +0200 (CEST)

Dear stata-list,
I have two problems concerning instrumental variables.
I'm estimating a pooled panel regressions. One of my exogenous variables is a time-variant variable. Since it is taken from a sruvey, I would ike to instrument it to cope with measurement error.
(1) In a 2sls specification, is it possible to regress the other exogeneous variables which are time-varying on the time-invariant survey variable? It seems a bit strange to me, due to the panel structure of the data.
(2) the original endogenous variable is censored at 0, hence I use intreg. Is it safe if I do the first stage by hand, save the predicted values and plug it into intreg fpr a second stage regressions?
Thank you in advance.

*   For searches and help try:

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