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: Rivtest p-values

From   Jessica Ellen Leight <>
Subject   st: Rivtest p-values
Date   Thu, 07 Apr 2011 13:48:07 -0400


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?

Jessica Leight
*   For searches and help try:

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