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]

Re: st: Rivtest p-values

From   Keith Finlay <>
Subject   Re: st: Rivtest p-values
Date   Thu, 7 Apr 2011 13:24:07 -0500

Hi Jessica,

Leandro Magnusson and I recently fixed a bug in -rivtest- when weights
are used. I'm not sure if this is causing the deviation, but it's
worth verifying. I've submitted the revision to the Stata Journal, but
in the meantime you can download the revision using "net from";.

Hope that helps,


Keith Finlay
Assistant Professor
Department of Economics
206 Tilton Hall
Tulane University
New Orleans, LA 70118-5698
504-865-5869 fax

On Thu, Apr 7, 2011 at 12:48 PM, Jessica Ellen Leight <> wrote:
> Hi,
> I am using rivtest to do weak instrument-robust inference following the use of
> the command ivreg2 with clustered standard errors.  I have found that for some
> equations, the p-values for the Anderson-Rubin test generated by rivtest are
> actually lower than those reported in ivreg2, such that a previously
> insignificant result becomes insignificant (and this is reflected in the
> confidence intervals as well.)   I was puzzled by this as I expected the
> opposite.  Is there any general rule about comparing p-values for conventional
> IV inference and those for weak-instrument robust inference, or something I
> should infer by virtue of the fact that the p-values drop?
> Best,
> Jessica Leight
> *
> *   For searches and help try:
> *
> *
> *

*   For searches and help try:

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