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: AW: ivreg2, xtivreg2 and FE IV.

From   "Martin Weiss" <>
To   <>
Subject   st: AW: ivreg2, xtivreg2 and FE IV.
Date   Thu, 22 Apr 2010 10:02:05 +0200


-help ivreg2- does not promise an "fe" option anywhere, does it? I
understand from various posts by its authors that -xtivreg2- is a wrapper
for -ivreg2- which does indeed provide an "fe" option, so yes, you should
install -xtivreg2- and see whether it does what you want.


-----Ursprüngliche Nachricht-----
[] Im Auftrag von Arjan Non
Gesendet: Donnerstag, 22. April 2010 09:41
Betreff: st: ivreg2, xtivreg2 and FE IV.

Dear Statalist

My aim is to estimate an instrumental regression in a fixed effects
panel model, where the standard errors should be corrected for
clustering on the fixed effects level. The standard xtivreg command
does not support cluster-robust standard errors. The command ivreg2
does, but this command produces weird results, in the sense that the
coefficient estimates differ a lot from the estimates produced by
ivreg and manual 2sls. Therefore, I decided to install the most recent
version, but now ivreg2 gives an error, saying that the option fe is
not allowed. Should I install xtivreg2 command? Does anyone know how
to solve this problem? I currently use Stata 11.

Thanks in advance

Arjan Non,
Phd Erasmus university Rotterdam
*   For searches and help try:

*   For searches and help try:

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