Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: Autocorrelation in panel data: xtregar or cluster() option ?
From
Simon Pickert <[email protected]>
To
[email protected]
Subject
st: Autocorrelation in panel data: xtregar or cluster() option ?
Date
Sun, 1 Dec 2013 22:53:12 +0100
Hi Statalisters,
I am looking for advice on how to account for auto-correlation in my unbalanced panel data set on which I want to run a fixed-effects regression. So far, I understand that there are two main possibilities:
xtreg with the cluster() option
or
xtregar
I have an unbalanced panel data set with 40K observations spread across 495 groups.
Min observations per group = 19
Max observations per group = 209
Avg observations per group = 82
How can I decide which method to use? The results are quite different in my case.
I found an interesting slide deck from Austin Nichols and Mark Schaffer. http://repec.org/usug2007/crse.pdf
It seems that the cluster() option performs best.
Thank you for your advice.
Simon
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/