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: How to determine bandwidth in ivreg2?

From   ling jin <>
Subject   st: How to determine bandwidth in ivreg2?
Date   Sun, 26 Jun 2011 14:42:12 +0800

Dear Statalisters,

I use "bw(#)" combined with "robust" in the command "ivreg2" to
address heteroskedasticity and autocorrelation. What's the criteria to
determine bandwidth?

A simple note for the choice of bandwidth I've read is "T^(1/3)" in
Baum, Schaffer and Stillman (2007). Do you know why? Is the rule used

Thank you so much!


Baum, Schaffer and Stillman (2007): Enhanced Routines for Instrumental
Variables/GMM Estimation and Testing
*   For searches and help try:

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