February 2010 - Dev - dynare.org

[Trac] #65: Second derivatives w.r.t. parameters
by Trac Dynare 15 Mar '10

15 Mar '10

#65: Second derivatives w.r.t. parameters --------------------------+------------------------------------------------- Reporter: ratto | Owner: sebastien Type: feature | Status: new Priority: major | Milestone: 4.2-alpha2 Component: Preprocessor | Version: Keywords: | --------------------------+------------------------------------------------- In identification analysis, for the computation of the Jacobian of typical optimization problems like moment matching (limited information). -- Ticket URL: <https://www.dynare.org/trac/ticket/65> Dynare <http://www.dynare.org> The Dynare project

1 5

[Trac] #32: Report MATLAB crashes on Linux/64 to Mathworks
by Trac Dynare 10 Mar '10

10 Mar '10

#32: Report MATLAB crashes on Linux/64 to Mathworks --------------------------+------------------------------------------------- Reporter: sebastien | Owner: Type: bug | Status: new Priority: major | Milestone: Component: Core M-files | Version: Keywords: | --------------------------+------------------------------------------------- On Linux, the 64-bits version of MATLAB crashes on some models with Dynare. The crash seems to occur with rather big models. It is reproducible, in the sense that the same code will always trigger the bug; but a small change in the code can sometimes make the crash disappear. Note that the crash is only triggered by M-files (no MEX). See the attached "plouf.m" and "plouf_static.m". They are derived from preprocessor-generated files with the EAGLE model. The script "plouf.m" calls the function "plouf_static.m". Attached is the crash dump of MATLAB. The crash is reproducible on MATLAB Linux/64, versions 7.5, 7.6 and 7.8. But it does not occur on MATLAB for Linux/32, Windows/32 and Windows/64. As can be seen from the displayed marks which I have inserted, MATLAB crashes around line 131 of "plouf_static.m". More precisely, it seems to crash just after having computed the expression there. I can't understand exactly what's going on. We need to report this to Mathworks. -- Ticket URL: <https://www.dynare.org/trac/ticket/32> Dynare <http://www.dynare.org> The Dynare project

1 3

[Trac] #60: Fix memory leaks in k_order_perturbation DLL
by Trac Dynare 09 Mar '10

09 Mar '10

#60: Fix memory leaks in k_order_perturbation DLL -------------------------+-------------------------------------------------- Reporter: sebastien | Owner: george Type: bug | Status: new Priority: minor | Milestone: Component: K-order DLL | Version: Keywords: | -------------------------+-------------------------------------------------- -- Ticket URL: <https://www.dynare.org/trac/ticket/60> Dynare <http://www.dynare.org> The Dynare project

1 6

Re: [DynareDev] [Trac] #9: Add derivatives of external functions in the model
by Trac Dynare 23 Feb '10

23 Feb '10

#9: Add derivatives of external functions in the model ---------------------------+------------------------------------------------ Reporter: sebastien | Owner: houtanb Type: feature | Status: assigned Priority: major | Milestone: 4.2 Component: Preprocessor | Version: Resolution: | Keywords: ---------------------------+------------------------------------------------ Comment(by sebastien): First version of this feature implemented in commit 12c4a52ebc5e95af537a7b7163bde2d34de2e46f. Still need to add: * inside DataTree, a table for maximum sharing between nodes associated to external functions * computation of temporary terms, in particular a way to optimize calls to jacobian and hessian functions -- Ticket URL: <https://www.dynare.org/trac/ticket/9#comment:6> Dynare <http://www.dynare.org> The Dynare project

1 0

[Trac] #84: normcdf() doesn't work with k_order
by Trac Dynare 23 Feb '10

23 Feb '10

#84: normcdf() doesn't work with k_order --------------------------+------------------------------------------------- Reporter: sebastien | Owner: sebastien Type: bug | Status: new Priority: minor | Milestone: Component: Preprocessor | Version: 4.1.0 Keywords: | --------------------------+------------------------------------------------- The reason is that normcdf() is not translated into C code. It can easily be fixed using the erf() function available from the math library. -- Ticket URL: <https://www.dynare.org/trac/ticket/84> Dynare <http://www.dynare.org> The Dynare project

1 1

[Trac] #89: Add bsxfun function for old versions of matlab.
by Trac Dynare 23 Feb '10

23 Feb '10

#89: Add bsxfun function for old versions of matlab. --------------------------+------------------------------------------------- Reporter: stepan | Owner: stepan Type: bug | Status: new Priority: major | Milestone: 4.2 Component: Core M-files | Version: Keywords: | --------------------------+------------------------------------------------- Built in function bsxfun appeared in matlab version 7.4. Dynare (with matlab's version < 7.4) may crash (depending on what the user is doing) because this function is missing. We have first to add an m file (in the missing subdirectory) doing the same job as bsxfun. This workaround will be at the cost of computing efficiency (built in bsxfun is very fast). So, if we really want to be compatible with all the versions of matlab since 7.0 (or 6.5.1 I can't remember) we have to write a mex file (it should be possible to use the source of the octave's built-in). -- Ticket URL: <https://www.dynare.org/trac/ticket/89> Dynare <http://www.dynare.org> The Dynare project

1 2

[Trac] #62: error message from histval syntax
by Trac Dynare 23 Feb '10

23 Feb '10

#62: error message from histval syntax --------------------------+------------------------------------------------- Reporter: michel | Owner: sebastien Type: bug | Status: new Priority: minor | Milestone: Component: Preprocessor | Version: Keywords: | --------------------------+------------------------------------------------- if one writes wrongly histval; y = 1; end; instead of histval; y(0)=1; end; there is no error message with the line number where the error occurs Note also that the reference manual is missleading because the (lag) doesn't appear in the syntax -- Ticket URL: <https://www.dynare.org/trac/ticket/62> Dynare <http://www.dynare.org> The Dynare project

1 3

[Trac] #90: Inconsistency between Brooks & Gelman convergence diagnostics and mh_drop
by Trac Dynare 18 Feb '10

18 Feb '10

#90: Inconsistency between Brooks & Gelman convergence diagnostics and mh_drop --------------------------+------------------------------------------------- Reporter: stepan | Owner: stepan Type: enhancement | Status: new Priority: trivial | Milestone: 4.2 Component: Core M-files | Version: Keywords: | --------------------------+------------------------------------------------- Brooks & Gelman convergence diagnostics are not always consistent with the way the posterior moments are computed. These diagnostics are built discarding half of the simulations at each step. By default the posterior moments are computed discarding half of the mcmc draws... But if the user changes the value of mh_drop, 100*mh_drop percent of the mcmc draws will be discarded when computing posterior moments, while the convergence diagnostics still discard the first half of the simulation at each step. The fix is to replace .5 by options_.mh_drop in McMCDiagnostics_core.m -- Ticket URL: <https://www.dynare.org/trac/ticket/90> Dynare <http://www.dynare.org> The Dynare project

1 1

[Trac] #81: bicgstab function does not exist in Octave 3.0
by Trac Dynare 17 Feb '10

17 Feb '10

#81: bicgstab function does not exist in Octave 3.0 --------------------------+------------------------------------------------- Reporter: sebastien | Owner: sebastien Type: bug | Status: new Priority: minor | Milestone: Component: Core M-files | Version: 4.1.0 Keywords: | --------------------------+------------------------------------------------- bicgstab function exists in both MATLAB and Octave 3.2, but not in Octave 3.0. This break option stack_solve_algo=3 under Octave 3.0. The fix is probably to backport the function from Octave 3.2 to Octave 3.0, and add in the matlab/missing/ subdir. -- Ticket URL: <https://www.dynare.org/trac/ticket/81> Dynare <http://www.dynare.org> The Dynare project

1 1

[Trac] #76: Under Octave/Windows, with USE_DLL, Dynare should not require "cygwin" or "msvc" option
by Trac Dynare 17 Feb '10

17 Feb '10

#76: Under Octave/Windows, with USE_DLL, Dynare should not require "cygwin" or "msvc" option --------------------------+------------------------------------------------- Reporter: sebastien | Owner: sebastien Type: bug | Status: new Priority: minor | Milestone: Component: Preprocessor | Version: 4.1.0 Keywords: | --------------------------+------------------------------------------------- The fix is easy: in ModFile.cc, use MATLAB's error() command instead of exiting the preprocessor when neither "cygwin" nor "msvc" is specified under MATLAB/Windows. -- Ticket URL: <https://www.dynare.org/trac/ticket/76> Dynare <http://www.dynare.org> The Dynare project

1 1