Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Re: Panel causality test


From   Kit Baum <baum@bc.edu>
To   asma esseghir <asmaesseghir@yahoo.com>
Subject   st: Re: Panel causality test
Date   Wed, 21 Apr 2010 12:21:08 -0400

Ah, this is one of the cases where  code just does not cope well with the data having been -tsset- as a panel. I once wrote a whole set of parallel utilities that would enable commands like official -dfuller- and -pperron- to work on a single panel (it is the option -onepanel- that is missing.  

-granger- is a very old chunk of code from STB-51. -findit granger- suggests using -gcause- from STB, whose help file says it should be used in preference to -granger-.  It does not work with panel data either) for reasons above, but I have addressed that. 

If you -ssc desc gcause- and install the package, routine -gcause2- will work on a single unit of a panel. E.g.

webuse grunfeld
gcause2 invest mvalue if company==4, lag(4)

will work.

Kit

On Apr 21, 2010, at 10:43 AM, asma esseghir wrote:

> Hi
> I would like to use the granger test but it is not appropriate with panel data? So what I have to do? Is there any other one or should I add something to the granger command for using panel data?
> 
> granger lngdp lneu
> 
> command may not be used with panel data
> 
> r(459); 
> 
> . granger lngdp lneu if pays==1
> 
> command may not be used with panel data
> 
> r(459);
> 
> thanks
> 
> 
> --- On Wed, 21/4/10, Kit Baum <baum@bc.edu> wrote:
> 
> From: Kit Baum <baum@bc.edu>
> Subject: Re: Panel causality test
> To: "asma esseghir" <asmaesseghir@yahoo.com>
> Cc: statalist@hsphsun2.harvard.edu
> Date: Wednesday, 21 April, 2010, 2:43 PM
> 
> 
> I'm not sure what test you're referring to. Do you mean you want to evaluate Granger causality for each unit in a panel, and somehow combine the results of those individual tests, as one would with a panel unit root test? Do you have a reference to the literature in which such a test is described?
> 
> On Apr 21, 2010, at 9:41 AM, asma esseghir wrote:
> 
> > Hi
> > Is there any panel causality test in stata? Because I didn’t find any one!
> > 
> > Thanks a lot for your help
> > 
> > Asma
> > 
> > 
> > 
> > 
> > 
> 
> 
> Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
>                               An Introduction to Stata Programming  |   http://www.stata-press.com/books/isp.html
>    An Introduction to Modern Econometrics Using Stata  |   http://www.stata-press.com/books/imeus.html
> 
> 


Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
                              An Introduction to Stata Programming  |   http://www.stata-press.com/books/isp.html
   An Introduction to Modern Econometrics Using Stata  |   http://www.stata-press.com/books/imeus.html


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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   |   Site index