Could someone have a look at the lines for setting the priors for correlation in set_prior.m (occurring in at least two places): ub = [ub; max(min(estim_params_.corrx(:,5),1),-1)]; lb = [lb; max(min(estim_params_.corrx(:,4),1),-1)]; In case estim_params_.corrx(:,4) is NaN, the lower bound defaults to +1 instead of -1. My conjecture is it should be lb = [lb; max(min(estim_params_.corrx(:,4),-1),-1)]; -- Johannes Pfeifer Haußerstr. 29 72076 Tübingen Tel.: +49-(0)7071-6396184 Mobil.: +49-(0)170-6936820 Germany
On 06/11/2012 11:15 PM, Johannes Pfeifer wrote:
Could someone have a look at the lines for setting the priors for correlation in set_prior.m (occurring in at least two places):
ub = [ub; max(min(estim_params_.corrx(:,5),1),-1)];
lb = [lb; max(min(estim_params_.corrx(:,4),1),-1)];
In case estim_params_.corrx(:,4) is NaN, the lower bound defaults to +1 instead of -1.
Good catch!
My conjecture is it should be
lb = [lb; max(min(estim_params_.corrx(:,4),-1),-1)];
No, this would always be -1 What about lb = [lb; min(max(estim_params_.corrx(:,4),-1),1)]; ? Best Michel
--
Johannes Pfeifer
Haußerstr. 29
72076 Tübingen
Tel.: +49-(0)7071-6396184
Mobil.: +49-(0)170-6936820
Germany
_______________________________________________ Dev mailing list Dev@dynare.org https://www.dynare.org/cgi-bin/mailman/listinfo/dev
On 13. Juni 2012 13:15, Michel Juillard wrote: What about lb = [lb; min(max(estim_params_.corrx(:,4),-1),1)]; Yes you are right, this needs to be changed for both ncx and ncn. -------- Johannes Pfeifer Haußerstr. 29 72076 Tübingen Tel.: +49-(0)7071-6396184 Mobil: +49-(0)170-6936820 jpfeifer@gmx.de Von: dev-bounces@dynare.org [mailto:dev-bounces@dynare.org] Im Auftrag von Michel Juillard Gesendet: Mittwoch, 13. Juni 2012 13:15 An: List for Dynare developers Betreff: Re: [DynareDev] set_prior.m On 06/11/2012 11:15 PM, Johannes Pfeifer wrote: Could someone have a look at the lines for setting the priors for correlation in set_prior.m (occurring in at least two places): ub = [ub; max(min(estim_params_.corrx(:,5),1),-1)]; lb = [lb; max(min(estim_params_.corrx(:,4),1),-1)]; In case estim_params_.corrx(:,4) is NaN, the lower bound defaults to +1 instead of -1. Good catch! My conjecture is it should be lb = [lb; max(min(estim_params_.corrx(:,4),-1),-1)]; No, this would always be -1? What about lb = [lb; min(max(estim_params_.corrx(:,4),-1),1)]; Best Michel -- Johannes Pfeifer Haußerstr. 29 72076 Tübingen Tel.: +49-(0)7071-6396184 Mobil.: +49-(0)170-6936820 Germany _______________________________________________ Dev mailing list Dev@dynare.org https://www.dynare.org/cgi-bin/mailman/listinfo/dev
Thanks Johannes, I will do it Best Michel On 06/13/2012 02:18 PM, Johannes Pfeifer wrote:
On 13. Juni 2012 13:15, Michel Juillard wrote:
What about lb = [lb; min(max(estim_params_.corrx(:,4),-1),1)];
Yes you are right, this needs to be changed for both ncx and ncn.
--------
Johannes Pfeifer
Haußerstr. 29
72076 Tübingen
Tel.: +49-(0)7071-6396184
Mobil: +49-(0)170-6936820
jpfeifer@gmx.de
*Von:*dev-bounces@dynare.org [mailto:dev-bounces@dynare.org] *Im Auftrag von *Michel Juillard *Gesendet:* Mittwoch, 13. Juni 2012 13:15 *An:* List for Dynare developers *Betreff:* Re: [DynareDev] set_prior.m
On 06/11/2012 11:15 PM, Johannes Pfeifer wrote:
Could someone have a look at the lines for setting the priors for correlation in set_prior.m (occurring in at least two places):
ub = [ub; max(min(estim_params_.corrx(:,5),1),-1)];
lb = [lb; max(min(estim_params_.corrx(:,4),1),-1)];
In case estim_params_.corrx(:,4) is NaN, the lower bound defaults to +1 instead of -1.
Good catch!
My conjecture is it should be
lb = [lb; max(min(estim_params_.corrx(:,4),-1),-1)];
No, this would always be -1?
What about lb = [lb; min(max(estim_params_.corrx(:,4),-1),1)];
Best
Michel
--
Johannes Pfeifer
Haußerstr. 29
72076 Tübingen
Tel.: +49-(0)7071-6396184
Mobil.: +49-(0)170-6936820
Germany
_______________________________________________ Dev mailing list Dev@dynare.org <mailto:Dev@dynare.org> https://www.dynare.org/cgi-bin/mailman/listinfo/dev
_______________________________________________ Dev mailing list Dev@dynare.org https://www.dynare.org/cgi-bin/mailman/listinfo/dev
participants (2)
-
Johannes Pfeifer -
Michel Juillard