I programmed something up last year. You can find it here:
http://www.wiso.uni-hamburg.de/fileadmin/sozialoekonomie/vwl/fritsche/Tarassow/var_girf_v104.inp

I think today I would have programmed it differently, and most likely more efficient, but as far as I remember II obtained very similar results to e.g. Eviews. Maybe it's not working without adjustment to some recent changes in the hansl language.
Have a look on it, and improve it if you find some issues.

Of course, you should also check out the Jack's SVAR package!

I don't remember the C-file and relevant code line you must adjust. You should have a look in the mailing list last year

Artur

2013/10/23 Logan Kelly <logan.kelly@uwrf.edu>

Thanks Arthur, how do I change the number of  non-convergent cases?  Or should I just code the bootstrap in hansel myself? 

--
Logan J. Kelly, Ph.D.
Assistant Professor, Department of Economics, University of Wisconsin-River Falls
Director, UWRF Center for Economic Research, http://www.uwrf.edu/cer
Section Editor, Global Business and Economics Review, http://www.inderscience.com/gber

23 D South Hall
College of Business and Economics
University of Wisconsin - River Falls
410 S. Third Street
River Falls, WI 54022-5001



-------- Original message --------
From: "Artur T." <artur.tarassow@googlemail.com>
Date: 10/23/2013 12:48 PM (GMT-06:00)
To: Gretl list <gretl-users@lists.wfu.edu>
Subject: Re: [Gretl-users] Confidence bounds on IRF


@Sven: I think he speaks about the native irf() command.

I had a similar issue in the past. We discussed this a while ago. The problem may occur if your data show excessive collinearity, for instance. If this happens more than 10 times within the 999 bootstrap replications, gretl breaks the loop and gives this error messages.

First, I circumenvented this issue by increasing the number of possible unstable cases in the C-code, before I programmed it myself. Alternatively, one could increase the numbers of non-convergent cases to, let's say, 99 or so.

Artur



2013/10/23 Sven Schreiber <svetosch@gmx.net>
Am 23.10.2013 18:40, schrieb Logan Kelly:
> I am using the latest snapshot 64 bit version on Win8 64bit.
>
> I am having an issue the irf function. I get the err Matrix is not
> positive definite if I ask for bootstrapped confidence bounds,

Are you talking about the SVAR package here?

-s
_______________________________________________
Gretl-users mailing list
Gretl-users@lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users


_______________________________________________
Gretl-users mailing list
Gretl-users@lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users