Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: durbin watson


From   Christopher Baum <baum@bc.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: durbin watson
Date   Fri, 05 May 2006 16:21:05 -0400

Rudy said
Does Stata have a way of calculating the p-value for a Durbin-Watson 
statistic? SAS does this and it is a lot easier for students because 
they don't have to rely on a Durbin-Watson table which can result in the 
test being inconclusive.

I would advise against reliance on the D-W statistic. IMHO the Breusch-Godfrey test is much more useful. When you consider a single lag, B-G is "Durbin's alternative test" that one uses when the D-W is inappropriate. But with multiple lags, one can consider higher-order AR or MA alternatives. Students all too often consider a satisfactory (or inconclusive) D-W as evidence of independence of the errors, a quite erroneous conclusion. I would be much more inclined to believe the results of bgodfrey with a healthy number of lags.

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