It seems to me that you should be able to do this (get impulse responses) 'by hand': 1) generate k-step ahead forecasts of your model with no shocks (ie, the unconditional mean) -> yf1 2) generate the same forecasts assuming a shock to a certain variable of a certain size -> yf2 3) IRF = yf2 - yf1
I haven't tried this myself in gretl (though I've done it in Excel), but it should be straightforward, no? PS ________________________________________ From: gretl-users-bounces(a)lists.wfu.edu [gretl-users-bounces(a)lists.wfu.edu] On Behalf Of Ricardo Gonçalves Silva [ricardogs(a)terra.com.br] Sent: Friday, May 29, 2009 6:02 PM To: Gretl list Subject: Re: [Gretl-users] VEC Question I also was wondering this point. I will use other software. Thanks in advance, Ricardo. -------------------------------------------------- From: "Sven Schreiber" <svetosch(a)gmx.net> Sent: Friday, May 29, 2009 7:11 PM To: "Gretl list" <gretl-users(a)lists.wfu.edu> Subject: Re: [Gretl-users] VEC Question > well impulse responses are of course a kind of forecast, so I'm afraid > that this won't solve your problem then... > > -sven > > Am 29.05.2009 17:51, Ricardo Gonçalves Silva schrieb: >> Hi Sven, >> >> Nice. I need the estimates and the impulses responses. >> >> Thanks >> >> Ricardo >> >> -------------------------------------------------- >> From: "Sven Schreiber" <svetosch(a)gmx.net> >> Sent: Friday, May 29, 2009 6:15 AM >> To: "Gretl list" <gretl-users(a)lists.wfu.edu> >> Subject: Re: [Gretl-users] VEC Question >> >>> Hi Ricardo, >>> >>> from the changelog of 1.8.1: >>> "VARs: support gaps in the lag structure (not for VECMs); >>> also, allow single-equations VARs" >>> >>> So for VECs it's not directly possible. Depending on what you want to do >>> with the model you could specify the 12th lag manually as an >>> (unrestricted) exogenous variable. Of course forecasts become tricky >>> then, but maybe you just need the estimates? >>> >>> cheers, >>> sven >>> >>> >>> Am 28.05.2009 18:12, Ricardo Gonçalves Silva schrieb: >>>> Hi, >>>> >>>> I'm estimating some VEC models and I would like to know if there is a >>>> way to choose specific lags for the model. >>>> Using the GUI, if I select 12 lags, GRETL uses lags 1 to 12; but I just >>>> want lag 12. >>>> Any help? >>>> Thanks >>>> >>>> Rick >>>> >>>> >>>> ------------------------------------------------------------------------ >>>> >>>> _______________________________________________ >>>> Gretl-users mailing list >>>> Gretl-users(a)lists.wfu.edu >>>> http://lists.wfu.edu/mailman/listinfo/gretl-users >>> >>> >>> _______________________________________________ >>> Gretl-users mailing list >>> Gretl-users(a)lists.wfu.edu >>> http://lists.wfu.edu/mailman/listinfo/gretl-users >>> >> >> >> >>> >>> Nenhum vírus encontrado nessa mensagem recebida. >>> Verificado por AVG - www.avgbrasil.com.br >>> Versão: 8.5.339 / Banco de dados de vírus: 270.12.45/2141 - Data de >>> Lançamento: 05/29/09 06:28:00 >>> >> >> _______________________________________________ >> Gretl-users mailing list >> Gretl-users(a)lists.wfu.edu >> http://lists.wfu.edu/mailman/listinfo/gretl-users >> > > > _______________________________________________ > Gretl-users mailing list > Gretl-users(a)lists.wfu.edu > http://lists.wfu.edu/mailman/listinfo/gretl-users > > > Nenhum vírus encontrado nessa mensagem recebida. > Verificado por AVG - www.avgbrasil.com.br > Versão: 8.5.339 / Banco de dados de vírus: 270.12.46/2142 - Data de > Lançamento: 05/29/09 17:53:00 > _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-users
