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: Shea's R2 with xtivreg2

From   "Schaffer, Mark E" <[email protected]>
To   <[email protected]>
Subject   st: RE: Shea's R2 with xtivreg2
Date   Sun, 3 Oct 2010 16:34:47 +0100


> -----Original Message-----
> From: [email protected] 
> [mailto:[email protected]] On Behalf Of 
> Erasmo Giambona
> Sent: 03 October 2010 15:32
> To: statalist
> Subject: st: Shea's R2 with xtivreg2
> Dear statalist,
> I am eastimating a model using xtivreg2 under Stata 10. I 
> noticed that the standard output no longer contains the 
> Shea's partial R2. Does anyone have any suggestion on how I 
> could make xtiver2 display it?

Shea's partial R2 is saved in the matrix macro e(first).  However,
there's no good reason to use it now that Angrist & Pischke have worked
out the "right" first-stage F statistic when there are multiple
endogenous regressors.  There is a discussion with references in the
-ivreg2- help file.  In a nutshell, unlike the Shea statistic, the A-P
first-stage statistics have known distributions and hence are readily


> Thanks,
> Erasmo
> *
> *   For searches and help try:
> *
> *
> *

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