Re: Re: Re: st:Normality and Granger Causality in Panel data

["Martin Schroeder" <schroeder_martin1@web.de>]

Dear Statalisters,
 
after having searched the Stata archives and consulted numerous text books for hours, I have not been able to obtain an answer (only references to the same question with counter questions) to the following simple question:
Does a Granger test make sense when it is performed with all units of a Panel dataset *at the same time* and if so, is there a Stata routine to run it?

Hood et al. 2008 (see reference below) do exactly that by comparing "the sum of squared residuals from a set of restricted models (RSS2) to the sum of squared residuals produced by a set of baseline (unrestricted) models (RSS1)" (p. 330). However, I can’t find a way to conduct such a test in Stata 12 with all units of a panel data set but only for each single unit (i.e. - var var1 var2 if unit==1 -, then - vargranger -; - gcause2 – doesn’t help here either).

See some of the older discussions in the archive:
http://www.stata.com/statalist/archive/2011-09/msg00932.html - Cameron's impressive reading list doesn't seem to contain any hint regarding a Stata routine that conducts a Granger test with all units contained in a panel data set.
http://www.stata.com/statalist/archive/2012-04/msg00158.html- "As for Granger causality, it's not clear how to think about that in a panel context." I take Kit’s comment that it is not appropriate after all? How come Hood et al. do exactly that?

Thanks for any hints,

Martin

 Hood, M.V., III, Kidd, Q., & Morris, I.L. (2008). Two Sides of the Same Coin? Employing Granger Causality Tests in a Time Series Cross-Section Framework. Political Analysis, 16, 324–344. http://pan.oxfordjournals.org/content/16/3/324.full.pdf

(just recently joined stata list, can't reply to Kit's old posting directly)
 

*
*   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/