Dear Kadir,

in version 4.2.3, we don't use the steady state Kalman filter. We only added it back to the unstable version. If GPM uses it, maybe it will provide a small speed gain. We will check

Best

Michel

On 11/30/2011 10:42 PM, Tanyeri, Kadir wrote:

Dear Michel,

Thank you very much. You are correct, I must have downloaded the unstable version.

But since it comes up while running our model, doesn’t it mean we would need it? Well maybe in the stable version it calls one of the other kalman filter programs…

Best,

Kadir

*From:*Michel Juillard [mailto:michel.juillard@mjui.fr] *Sent:* Wednesday, November 30, 2011 4:35 PM *To:* List for Dynare developers *Cc:* Sébastien Villemot; Tanyeri, Kadir *Subject:* Re: [DynareDev] [GPM Maintainers] Fwd: Release of Dynare 4.2.3

Dear Kadir,

thank you for reporting the problem. You are right. What is strange is that univariate_kalman_filter_ss.m exist only in unstable but not in 4.2.3

All the best,

Michel

On 11/30/2011 09:06 PM, Sébastien Villemot wrote:

Hi Kadir,

I forward this to the Dynare developers' list, since I’m not familiar with this part of the code.

Best,

"Tanyeri, Kadir"KTanyeri@imf.org mailto:KTanyeri@imf.org writes:

Sebastien,



Thank you very much for the much anticipated version!



There seems to be small typo in the following file around line 100, I think it should be as the second line below:



..\4.2.3\matlab\kalman\likelihood\univariate_kalman_filter_ss.m



%Ki = (PP*Z(i,:))'/Fi;

  Ki = (PP*Z(i,:)')/Fi;



Could you please check?



Best,

Kadir





-----Original Message-----

From:gpm-maintainers-bounces@cepremap.org  <mailto:gpm-maintainers-bounces@cepremap.org>  [mailto:gpm-maintainers-bounces@cepremap.org] On Behalf Of Sébastien Villemot

Sent: Wednesday, November 30, 2011 12:16 PM

To: GPM Maintainers

Subject: [GPM Maintainers] Fwd: [DynareDev] Release of Dynare 4.2.3



Hi,



We made a new Dynare release, which hopefully should fix the issues you recently encountered. We were able to estimate GPM6 with this new version.



Please tell us if you still encounter any problem.



Best,

---

Dev mailing list Dev@dynare.org mailto:Dev@dynare.org https://www.dynare.org/cgi-bin/mailman/listinfo/dev