Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: mean centering and FGLS


From   Kit Baum <baum@bc.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: mean centering and FGLS
Date   Fri, 19 Aug 2005 19:57:26 -0400

Steve said
I have always mean centered data when concerned with autocorrelation
with standard OLS regression.

I have not seen anything with regard to mean centering with FGLS but
should I do so before running data?



SInce altering the means of Y and X by a constant only affects the intercept in regression (and centering sets it to zero) it has nothing whatsoever to do with the serial correlation of the disturbance process. Centering the data merely causes the regression surface to pass through (0,0 [0,0,0,...]). It will have no effect on, say, a consistent estimate of \rho in an AR(1) error scheme.

Kit Baum, Boston College Economics
http://ideas.repec.org/e/pba1.html


*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/




© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index