Statalist


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

Re: st: Autocorrelation in FE and pantest2 problems


From   nicola.baldini2@unibo.it
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Autocorrelation in FE and pantest2 problems
Date   Sun, 12 Aug 2007 15:10:19 +0200

-pantest2- (remember the dashes!) works perfectly on Intercooled Stata 9.2 fully updated.
Dynamics methods usually report a test for autocorrelation. Unfortunately, they are random effects methods.
Given that T is almost equal to N, my suggestion is to use -xtpmg- from SJ7-2 (you can also test if autocorrelation is really a problem). 
 
Nicola

At 02.33 10/08/2007 -0400, John Bunge wrote:
>Dear Users.
>
>I want to implement a panel data regression on a T=20 and N=24 Macro Panel.
>
>First I run FE. Then I want to check for autocorrelation in the residuals in order to have a criterion if dynamic methods (anderson/hsiao, arellano/bond, blundell/bond) should be applied due to a lagged dependent variable.
>
>Related to this are two questions:
>
>1) As a serial correlation test for the residuals of a FE regression I only know the command pantest2. Are there any other commands? I read about the LM test for 1st order serial correlation in a FE model in Baltagi's panel book (3rd ed., 2005), pp. 97-98. Is there a command for it in Stata?
>
>2) If I apply the command pantest2 the order in my dataset was changed sometimes (the order of observation units remains but the time order within those units change in a apparently random fashion), so pantest2 does not run. Thats really strange since I always have tsset my data! Is this a general problem with this command or may it be caused by my Stata Version?
>
>Best regards,
>
>John. 
*
*   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