Dynare forums

Hi, I'm trying estimate a DSGE model using baeysian approach. Since octave don't read excel files, I'm trying to work with a mat-file; recordind the datas in a txt-file and them saving as a .mat-file. But when I try estimate the model, the follow message appears: ...

... model derivatives: - order 1 Computing dynamic model derivatives: - order 1 Processing outputs ...done Preprocessing completed. Starting MATLAB/Octave computing. Residuals of the static equations: Equation number 1 : 0 Equation number 2 : 0 Equation number 3 : 0 Equation number 4 : 0 Equation ...

Thanks for reporting this problem and for suggesting a fix.

Your fix has been incorporated in the unstable version of Dynare, and will be incorporated in the next stable release (4.2.2).

Best,

Dear Kyriacos

I never tried... It should work, if I believe the specs given by Mathworks.

If you cannot afford Matab you can give a try to Octave (a free clone of Matlab).

Best,
Stéphane.

... model derivatives: - order 1 Computing dynamic model derivatives: - order 1 Processing outputs ...done Preprocessing completed. Starting MATLAB/Octave computing. STEADY-STATE RESULTS: Y 0.00625684 C 0.0185632 I -0.0123064 K -0.492254 Ke 0.0143672 D 0.137502 L 0.644124 H 0.377415 He 0.0759695 ...

... - order 1 Computing dynamic model derivatives: - order 1 - order 2 Processing outputs ...done Preprocessing completed. Starting MATLAB/Octave computing. SOLVE: maxit has been reached ??? Error using ==> steady_ at 132 STEADY: convergence problems. Thanks to everyone who helps me. best ...

Try to debug the problem with Dynare Matlab/Octave. It seems to arise in computing the steady state under optimal policy.

Best

Michel

I don't think it is possible. Try to compute the steady state with Dynare Matlab/Octave and provide better initial values to Dynare++. Note also that Dynare Matlab/Octave provides now 3rd order approximation.

Best

Michel

Thank you. I very much appreciate the update!

This is a bug either with Octave or Dynare. Will find the source (and fix if it's Dynare) and get back to you.

I use Octave version 3.2.4 together with GUI Octave 1.3.0. I run Dynare 4.2.1 and use the program RBC_schat.mod, with RBC_schat_steadystate.m and the data file simRBCgrow.m. The OS is Windows XP Professions version 2002 service ...

What OS & Octave version are you running?

Also, please provide a mod file that reproduces this error.

I am unable to open or print the eps graphics files made by Octave after performing a bayesian estimation. Has anyone else had this problem?

Sincerely,
Rob Luginbuhl

I've now got Octave running by changing line 204 in the file metropolis_hastings_initialization.m to read

files = dir ( strcat ( MhDirectoryName, filesep, ModelName, '_mh*.mat') );

I hope this helps.

Sincerely,
Rob

I have tried to use the arguments mode=0 and load_mh to increase the length of the mh chain when estimating the RBC model using Octave (this works fine in Matlab), but I obtain the following error message. >> dynare RBC_schat Configuring Dynare ... [mex] Generalized QZ. [mex] Sylvester ...