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: fixed effects with multicollinearity

From   "Reddy, Colin" <>
To   "" <>
Subject   st: fixed effects with multicollinearity
Date   Fri, 29 Jul 2011 08:04:18 +0000

Dear Listers

I have variables that have high multicollinearity due to interaction terms. Do I have to mean centre these explanatory variables (or all of them, even the controls) even though the within fe estimator does mean centering automatically?



This email and all contents are subject to the following disclaimer:

*   For searches and help try:

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