dyn_second_order_solver.m
Dear Michel, I just fixed a bug in dyn_second_order_solver.m and pushed it to my repository. Could you please commit it and also port it to 4.3 Could you also please have a look at the file. Line 198 reads zud=[zeros(np,M_.exo_det_nbr);dr.ghud{1};gx(:,1:npred)*hud;zeros(M_.exo_nbr, M_.exo_det_nbr);eye(M_.exo_det_nbr)]; Matlab gives me a warning that gx is used before it is defined. I think this is another bug. Finally, there are either missing parts or the file needs cleaning. Several variables seem to be defined, but are never used: kad, kae, n2, n4, k0 in line 86, offset, kk in line 169, Gu and Guu in lines 172-173 Best, Johannes -- Johannes Pfeifer Haußerstr. 29 72076 Tübingen Tel.: +49-(0)7071-6396184 Mobil.: +49-(0)170-6936820 Germany
"Johannes Pfeifer" <jpfeifer@gmx.de> writes:
I just fixed a bug in dyn_second_order_solver.m and pushed it to my repository. Could you please commit it and also port it to 4.3
Fix pushed, thanks. I am not sure however of the exact impact of the bug. On a simple example (example1.mod) it seems to have no effect.
Could you also please have a look at the file. Line 198 reads
zud=[zeros(np,M_.exo_det_nbr);dr.ghud{1};gx(:,1:npred)*hud;zeros (M_.exo_nbr,M_.exo_det_nbr);eye(M_.exo_det_nbr)];
Matlab gives me a warning that gx is used before it is defined. I think this is another bug.
This is obviously a bug which affects exogenous deterministic variables at order 2. I don't know how to fix this, I leave it to Michel. I modified tests/example1_varexo_det.mod so that it tests 2nd order, in order not to forget this issue.
Finally, there are either missing parts or the file needs cleaning. Several variables seem to be defined, but are never used:
kad, kae, n2, n4, k0 in line 86, offset, kk in line 169, Gu and Guu in lines 172-173
I pushed a cleanup commit. Thanks, -- Sébastien Villemot Researcher in Economics http://www.dynare.org/sebastien Phone: +33-1-40-77-84-04
Thanks, the bug should have only led to ghuu not being saved in oo_.dr at second order. As the results were simply missing, there was no crash or other noticeable effect. -- Johannes Pfeifer Haußerstr. 29 72076 Tübingen Tel.: +49-(0)7071-6396184 Mobil.: +49-(0)170-6936820 Germany -----Ursprüngliche Nachricht----- Von: dev-bounces@dynare.org [mailto:dev-bounces@dynare.org] Im Auftrag von Sébastien Villemot Gesendet: Mittwoch, 19. September 2012 10:51 An: List for Dynare developers Betreff: Re: [DynareDev] dyn_second_order_solver.m "Johannes Pfeifer" <jpfeifer@gmx.de> writes:
I just fixed a bug in dyn_second_order_solver.m and pushed it to my repository. Could you please commit it and also port it to 4.3
Fix pushed, thanks. I am not sure however of the exact impact of the bug. On a simple example (example1.mod) it seems to have no effect.
Could you also please have a look at the file. Line 198 reads
zud=[zeros(np,M_.exo_det_nbr);dr.ghud{1};gx(:,1:npred)*hud;zeros (M_.exo_nbr,M_.exo_det_nbr);eye(M_.exo_det_nbr)];
Matlab gives me a warning that gx is used before it is defined. I think this is another bug.
This is obviously a bug which affects exogenous deterministic variables at order 2. I don't know how to fix this, I leave it to Michel. I modified tests/example1_varexo_det.mod so that it tests 2nd order, in order not to forget this issue.
Finally, there are either missing parts or the file needs cleaning. Several variables seem to be defined, but are never used:
kad, kae, n2, n4, k0 in line 86, offset, kk in line 169, Gu and Guu in lines 172-173
I pushed a cleanup commit. Thanks, -- Sébastien Villemot Researcher in Economics http://www.dynare.org/sebastien Phone: +33-1-40-77-84-04
Hi, in fact, it was only that, for models with auxiliary variables, the derivatives ghuu were saved with too many rows. That is why it took time for the bug to pop up. Good catch!. I look into the problem with exogenous deterministic variables. Best Michel On 09/19/2012 11:33 AM, Johannes Pfeifer wrote:
Thanks, the bug should have only led to ghuu not being saved in oo_.dr at second order. As the results were simply missing, there was no crash or other noticeable effect.
-- Johannes Pfeifer Haußerstr. 29 72076 Tübingen Tel.: +49-(0)7071-6396184 Mobil.: +49-(0)170-6936820 Germany
-----Ursprüngliche Nachricht----- Von: dev-bounces@dynare.org [mailto:dev-bounces@dynare.org] Im Auftrag von Sébastien Villemot Gesendet: Mittwoch, 19. September 2012 10:51 An: List for Dynare developers Betreff: Re: [DynareDev] dyn_second_order_solver.m
"Johannes Pfeifer"<jpfeifer@gmx.de> writes:
I just fixed a bug in dyn_second_order_solver.m and pushed it to my repository. Could you please commit it and also port it to 4.3 Fix pushed, thanks.
I am not sure however of the exact impact of the bug. On a simple example (example1.mod) it seems to have no effect.
Could you also please have a look at the file. Line 198 reads
zud=[zeros(np,M_.exo_det_nbr);dr.ghud{1};gx(:,1:npred)*hud;zeros (M_.exo_nbr,M_.exo_det_nbr);eye(M_.exo_det_nbr)];
Matlab gives me a warning that gx is used before it is defined. I think this is another bug. This is obviously a bug which affects exogenous deterministic variables at order 2. I don't know how to fix this, I leave it to Michel. I modified tests/example1_varexo_det.mod so that it tests 2nd order, in order not to forget this issue.
Finally, there are either missing parts or the file needs cleaning. Several variables seem to be defined, but are never used:
kad, kae, n2, n4, k0 in line 86, offset, kk in line 169, Gu and Guu in lines 172-173 I pushed a cleanup commit.
Thanks,
-- Sébastien Villemot Researcher in Economics http://www.dynare.org/sebastien Phone: +33-1-40-77-84-04
_______________________________________________ Dev mailing list Dev@dynare.org https://www.dynare.org/cgi-bin/mailman/listinfo/dev
Michel Juillard <michel.juillard@mjui.fr> writes:
in fact, it was only that, for models with auxiliary variables, the derivatives ghuu were saved with too many rows.
Are you sure? M_.endo_nbr contains the number of variables *including* the auxiliary… -- Sébastien Villemot Researcher in Economics http://www.dynare.org/sebastien Phone: +33-1-40-77-84-04
On 09/20/2012 05:19 PM, Sébastien Villemot wrote:
Michel Juillard<michel.juillard@mjui.fr> writes:
in fact, it was only that, for models with auxiliary variables, the derivatives ghuu were saved with too many rows. Are you sure? M_.endo_nbr contains the number of variables *including* the auxiliaryâEUR¦ You're right. Maybe this line is now useless. Johanhes, do you have a *.mod file that triggers the bug that you found?
Best Michel
_______________________________________________ Dev mailing list Dev@dynare.org https://www.dynare.org/cgi-bin/mailman/listinfo/dev
No. I don't. A friend of mine just noted the obvious typo. It is conceivable it had no consequences. - Johannes Pfeifer Department of Economics University of Tuebingen Mohlstr. 36 72074 Tuebingen Germany Am 21.09.2012 um 12:18 schrieb Michel Juillard <michel.juillard@mjui.fr>:
On 09/20/2012 05:19 PM, Sébastien Villemot wrote:
Michel Juillard <michel.juillard@mjui.fr> writes:
in fact, it was only that, for models with auxiliary variables, the derivatives ghuu were saved with too many rows. Are you sure? M_.endo_nbr contains the number of variables *including* the auxiliary…
You're right. Maybe this line is now useless. Johanhes, do you have a *.mod file that triggers the bug that you found?
Best
Michel
_______________________________________________ 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
Thanks Johannes Best Michel On 09/21/2012 12:54 PM, Johannes Pfeifer wrote:
No. I don't. A friend of mine just noted the obvious typo. It is conceivable it had no consequences.
- Johannes Pfeifer Department of Economics University of Tuebingen Mohlstr. 36 72074 Tuebingen Germany
Am 21.09.2012 um 12:18 schrieb Michel Juillard <michel.juillard@mjui.fr <mailto:michel.juillard@mjui.fr>>:
On 09/20/2012 05:19 PM, Sébastien Villemot wrote:
Michel Juillard<michel.juillard@mjui.fr> writes:
in fact, it was only that, for models with auxiliary variables, the derivatives ghuu were saved with too many rows. Are you sure? M_.endo_nbr contains the number of variables *including* the auxiliaryâEUR¦ You're right. Maybe this line is now useless. Johanhes, do you have a *.mod file that triggers the bug that you found?
Best
Michel
_______________________________________________ Dev mailing list Dev@dynare.org https://www.dynare.org/cgi-bin/mailman/listinfo/dev
_______________________________________________ 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
"Johannes Pfeifer " <jpfeifer@gmx.de> writes:
Thanks, the bug should have only led to ghuu not being saved in oo_.dr at second order. As the results were simply missing, there was no crash or other noticeable effect.
Actually the bug does not prevent oo_.dr.ghuu from being saved. I don't really understand what the faulty line is for.
-- Johannes Pfeifer Haußerstr. 29 72076 Tübingen Tel.: +49-(0)7071-6396184 Mobil.: +49-(0)170-6936820 Germany
-----Ursprüngliche Nachricht----- Von: dev-bounces@dynare.org [mailto:dev-bounces@dynare.org] Im Auftrag von Sébastien Villemot Gesendet: Mittwoch, 19. September 2012 10:51 An: List for Dynare developers Betreff: Re: [DynareDev] dyn_second_order_solver.m
"Johannes Pfeifer" <jpfeifer@gmx.de> writes:
I just fixed a bug in dyn_second_order_solver.m and pushed it to my repository. Could you please commit it and also port it to 4.3
Fix pushed, thanks.
I am not sure however of the exact impact of the bug. On a simple example (example1.mod) it seems to have no effect.
Could you also please have a look at the file. Line 198 reads
zud=[zeros(np,M_.exo_det_nbr);dr.ghud{1};gx(:,1:npred)*hud;zeros (M_.exo_nbr,M_.exo_det_nbr);eye(M_.exo_det_nbr)];
Matlab gives me a warning that gx is used before it is defined. I think this is another bug.
This is obviously a bug which affects exogenous deterministic variables at order 2. I don't know how to fix this, I leave it to Michel. I modified tests/example1_varexo_det.mod so that it tests 2nd order, in order not to forget this issue.
Finally, there are either missing parts or the file needs cleaning. Several variables seem to be defined, but are never used:
kad, kae, n2, n4, k0 in line 86, offset, kk in line 169, Gu and Guu in lines 172-173
I pushed a cleanup commit.
Thanks,
-- Sébastien Villemot Researcher in Economics http://www.dynare.org/sebastien Phone: +33-1-40-77-84-04
_______________________________________________ Dev mailing list Dev@dynare.org https://www.dynare.org/cgi-bin/mailman/listinfo/dev
-- Sébastien Villemot Researcher in Economics http://www.dynare.org/sebastien Phone: +33-1-40-77-84-04
Dynare saved dr.ghuu = A\rhs; while it should have saved dr.ghuu = dr.ghuu(1:M_.endo_nbr,:); where the RHS makes use of the previous result, i.e. only some rows are selected to be saved. I guess ghuu just contained too many rows so that the dimensions were not conformable. -- Johannes Pfeifer Haußerstr. 29 72076 Tübingen Tel.: +49-(0)7071-6396184 Mobil.: +49-(0)170-6936820 Germany -----Ursprüngliche Nachricht----- Von: dev-bounces@dynare.org [mailto:dev-bounces@dynare.org] Im Auftrag von Sébastien Villemot Gesendet: Mittwoch, 19. September 2012 15:13 An: List for Dynare developers Betreff: Re: [DynareDev] dyn_second_order_solver.m "Johannes Pfeifer " <jpfeifer@gmx.de> writes:
Thanks, the bug should have only led to ghuu not being saved in oo_.dr at second order. As the results were simply missing, there was no crash or other noticeable effect.
Actually the bug does not prevent oo_.dr.ghuu from being saved. I don't really understand what the faulty line is for.
-- Johannes Pfeifer Haußerstr. 29 72076 Tübingen Tel.: +49-(0)7071-6396184 Mobil.: +49-(0)170-6936820 Germany
-----Ursprüngliche Nachricht----- Von: dev-bounces@dynare.org [mailto:dev-bounces@dynare.org] Im Auftrag von Sébastien Villemot Gesendet: Mittwoch, 19. September 2012 10:51 An: List for Dynare developers Betreff: Re: [DynareDev] dyn_second_order_solver.m
"Johannes Pfeifer" <jpfeifer@gmx.de> writes:
I just fixed a bug in dyn_second_order_solver.m and pushed it to my repository. Could you please commit it and also port it to 4.3
Fix pushed, thanks.
I am not sure however of the exact impact of the bug. On a simple example (example1.mod) it seems to have no effect.
Could you also please have a look at the file. Line 198 reads
zud=[zeros(np,M_.exo_det_nbr);dr.ghud{1};gx(:,1:npred)*hud;zeros (M_.exo_nbr,M_.exo_det_nbr);eye(M_.exo_det_nbr)];
Matlab gives me a warning that gx is used before it is defined. I think this is another bug.
This is obviously a bug which affects exogenous deterministic variables at order 2. I don't know how to fix this, I leave it to Michel. I modified tests/example1_varexo_det.mod so that it tests 2nd order, in order not to forget this issue.
Finally, there are either missing parts or the file needs cleaning. Several variables seem to be defined, but are never used:
kad, kae, n2, n4, k0 in line 86, offset, kk in line 169, Gu and Guu in lines 172-173
I pushed a cleanup commit.
Thanks,
-- Sébastien Villemot Researcher in Economics http://www.dynare.org/sebastien Phone: +33-1-40-77-84-04
_______________________________________________ Dev mailing list Dev@dynare.org https://www.dynare.org/cgi-bin/mailman/listinfo/dev
-- Sébastien Villemot Researcher in Economics http://www.dynare.org/sebastien Phone: +33-1-40-77-84-04
participants (3)
-
Johannes Pfeifer -
Michel Juillard -
Sébastien Villemot