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

st: Re: test for AR in panel data


From   Christopher F Baum <baum@bc.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: Re: test for AR in panel data
Date   Tue, 3 Feb 2004 07:03:21 -0500

On Feb 3, 2004, at 2:33 AM, Kalina wrote:


I am not able to use the dwstat or bgtest tests in Stata 7 with a panel
data. Is there a fix for this problem or other tests I can use instead?

Yes (sort of). 'findit panelauto' will locate a set of minor modifications to official commands that I wrote. These routines will work on ONE timeseries of a panel: that is, you may analyze the residuals of unit i alone (presumably with some sort of loop over i). This is NOT a "panel durbin watson test', if you consider such a test one that gives you a single verdict on the panel, but one that removes the annoying restriction that many of Stata's official time-series commands (including unit root tests) will not work on a panel unless you drop all but one timeseries of the panel.

Also please note that 'bgtest' and 'durbinh' (Baum and Wiggins, STB-55) are now implemented in official Stata as 'bgodfrey' and 'durbina'. The latter routines are to be preferred, since you can consult StataCorp tech support for assistance with them.

KIt

*
* 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