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: robust clustered errors in panel data

From   "Anat (Manes) Tchetchik" <>
Subject   st: robust clustered errors in panel data
Date   Wed, 25 Dec 2013 17:05:19 +0200

Dear all,
I have a unbalanced panel of macroeconomic data of 35 countries over
17 years. I ran a fixed-effects model model where the effects are the
countries with robust standard errors.
A  reviewer suggested the use of clustered errors.
The question is do the clustering variable ought be the country_code
(same as the "effects") or can I use a different clustering variable
such as the sub-region the country belongs to (e.g. Latin America,
North America, West EU etc.).
If both clustering are plausible, then what should should be the
criterion of choosing?
I'll appreciate any advice,
Happy Christmas

Anat Tchetchik, PhD
Department of Business Administration
Guilford Glazer Faculty of Business and Management
Ben-Gurion University of the Negev
P.O.Box: 653
Beer-Sheva, Israel, 84105

Phone         972-(0)8-6479735
Fax:           972-(0)8-6472920
*   For searches and help try:

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