Re: [Gretl-users] Granger Causality Test

A "Granger test" entry in Gretl menu would be great ;-) Best, Henrique 2010/8/6 John C Frain <frainj(a)gmail.com <mailto:frainj@gmail.com>> In the example here one can conclude that PRICE does not Granger cause QNC. Gretl then gives the results of all bivariate Granger Causality tests. There are other possible Granger Causality tests. If one divides the n dimensional VAR into two sub-systems of dimension n1 and n2 one may ask if the first system Granger causes the second. In the example one may ask if PRICE Granger causes QNC and INCOME. This is a joint test on the lags of PRICE in the QNC and INCOME equations. AS far as I know this can not be implemented within the VAR estimation in the GUI. (I presume that the test can be implemented using the simultaneous equations estimation routines). Would it be possible or worthwhile to add these items to the test menu in the Vector autoregression results window. On 6 August 2010 20:17, Allin Cottrell <cottrell(a)wfu.edu <mailto:cottrell@wfu.edu>> wrote: > On Thu, 5 Aug 2010, Sven Schreiber wrote: > >> >> Am 05.08.2010 18:08, schrieb Carlos Manuel Gratereaux Hernández: >> > I am a teacher of Basic Econometric at UASD of Dominican Repúblic. Last >> > class I was working about Granger Causality Test with my students. I had >> > to use another econometric software because i didn't find this test in >> > windows language GRETL. How can I do it in GRETL? >> >> IIRC it appears automatically in the output if you estimate the >> corresponding VAR. Some time ago the buzzword "Granger causality" did >> not appear in the output, but it was just that test. An explanatory >> remark may have been added in the meantime, I don't have time right now >> to verify. > > No, it hasn't been added. The current output looks like this > (for an equation where QNC is the dependent variable): > > F-tests of zero restrictions: > > All lags of QNC F(4, 47) = 14.868 [0.0000] > All lags of PRICE F(4, 47) = 2.0829 [0.0980] > All lags of INCOME F(4, 47) = 4.3412 [0.0045] > All vars, lag 4 F(3, 47) = 11.755 [0.0000] > > The middle two F-tests (only) are Granger causality tests. Putting > the buzzword in would require some reorganization. Dos anyone have > a suggestion (if it's worth doing)? > > Allin Cottrell > > _______________________________________________ > Gretl-users mailing list > Gretl-users(a)lists.wfu.edu <mailto:Gretl-users@lists.wfu.edu> > http://lists.wfu.edu/mailman/listinfo/gretl-users > -- John C Frain Economics Department Trinity College Dublin Dublin 2 Ireland www.tcd.ie/Economics/staff/frainj/home.html <http://www.tcd.ie/Economics/staff/frainj/home.html> mailto:frainj@tcd.ie <mailto:frainj@tcd.ie> mailto:frainj@gmail.com <mailto:frainj@gmail.com> _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu <mailto:Gretl-users@lists.wfu.edu> http://lists.wfu.edu/mailman/listinfo/gretl-users -- Henrique C. de Andrade Doutorando em Economia Aplicada Universidade Federal do Rio Grande do Sul www.ufrgs.br/ppge <http://www.ufrgs.br/ppge> _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-users