Dear Michel
Thanks
I however think we should keep the original definition of four xxx_id but add four new ones for indices into Q and H on the lines of ncn_id1_id2, i.e. 2 more for the Q&H diagonals in addition to ncn_id1_id2 and ncx_id1_id2 for correlations.
E..g. lets call the new ones nvx_Q_id and nvn_H_id (and poss. rename ncn_id1_id2 and ncx_id1_id2 to ncn_H_id1_id2 and ncx_Q_id1_id2)
I.e.: by adding a re- definition of nvx_id and nvn_id in *Remark 5 * we lost vectors indices into xparam1 for Q and H elements which were covered by logic of Remark 4 applied to nvx_id and nvn_id and we lost ncx_id and ncn_id indices into xparam1 *i.e.
Old Remark 4.* xxx_id vectors select subset of xparam1 (for instance, xparam_1(np_id) selects all the lines of xparam1 corresponding to the estimated "deep" parameters).
could extend to nvx_id selecting estimated structural shock standard deviations form xparam1 but we now do not have such index..
PS: Also, *Remark 5 defines *nvx_id and nvn_id vectors but in example (for instance Q(*ncx_id,ncx_id*)) uses undefined *ncx_id* vector,
PS 2 I can make changes to the wiki if you agree.
Best regards George
On 28/04/2010 15:30, Michel Juillard wrote:
George,
I update the wiki page on this point. I hope it clarifies things.
Best
Michel
George Perendia wrote:
Hi Michel
As I understand, according to
"Remark 4.* xxx_id vectors select subset of xparam1 (for instance, xparam_1(np_id) selects all the lines of xparam1 corresponding to the estimated "deep" parameters"
on the wiki page, ncx_id is index into varexo1, so it can be just one column. but what we seem to be missing there are both single (for diagonals) and dual (for correlations) indices into Q and H matrices such as estim_params_.corrx(i,1). Best George On 28/04/2010 14:42, Michel Juillard wrote:
Hi George,
ncx_id ncx by 1 vector of integers.(for estimated structural shock correlations). .
ncn_id ncn by 1 vector of integers (estimated measurement error correlations).
these should be two-column matrices (one index of each correlated variable) ncx_id1_id2 ncx x 2 ncn_id1_id2 ncn x 2
Best
Michel
We have those *_id's indices into xparam1 but what indices are there into Q and H matrices respectively , including for pairs of (symmetric) coordinates on the lines of
estim_params_.corrx(i,1);and k1 = options_.lgyidx2varobs(estim_params_.corrn(i,1)); *used in DsgeLikelihood.m*?
Shall we continue to use those as separate entities in estim_params_, including. estim_params_.var_exo and estim_params_.var_endo, or store separate new ones into the new EstimationModule params structure?
-- Best regards George
Dev mailing list 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
--
Best regards George
Dev mailing list 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