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: RE: which F-test do I report from xtivreg first stage

From   "Schaffer, Mark E" <>
To   "" <>
Subject   st: RE: which F-test do I report from xtivreg first stage
Date   Fri, 6 Dec 2013 23:21:17 +0000


> -----Original Message-----
> From: [mailto:owner-
>] On Behalf Of David Torres
> Sent: 06 December 2013 21:32
> To:
> Subject: st: which F-test do I report from xtivreg first stage
> Hello again, Statalisters,
> I am utilizing the xtivreg's postestimation command, xtoverid (with the noisily
> option), to get first stage estimates along with the F-test.
> I am unsure which of the F-tests I'm supposed to report.  At the top of the OLS
> regression of the first stage there is the F for the model.

That's not the right one - it tests the joint significance of all the regressors, whereas you want the F that is a test of just the excluded instruments.

>  Below the first stage
> model, though, is the F-test of excluded instruments, also called the Angrist-
> Pischke multivariate F test of excluded instruments.

These are two different Fs unless you have a single endogenous regressor, in which case they coincide.  Sounds like they coincide in your case. 

>  I'm thinking it's the latter
> that I should be reporting.  Am I right?

Yes, that's the one.


> Diego
> *
> *   For searches and help try:
> *
> *
> *

Sunday Times Scottish University of the Year 2011-2013
Top in the UK for student experience
Fourth university in the UK and top in Scotland (National Student Survey 2012)

We invite research leaders and ambitious early career researchers to 
join us in leading and driving research in key inter-disciplinary themes. 
Please see for further information and how
to apply.

Heriot-Watt University is a Scottish charity
registered under charity number SC000278.

*   For searches and help try:

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