Possible Bug: Dynare 4.3.1 and Wieland's macromodel database

[vvd]

Hello,

I was wondering if anybody ran into the similar problem. On my computer, Wieland's MMB runs fine under Matlab 7.10.0 & Dynare 4.2.5 (Windows 32-bit).
However, it does not with Dynare 4.3.1, complaining that

??? Input argument "oo" is undefined.

Below is the complete error statement I receive.

Starting Dynare (version 4.3.1).
Starting preprocessing of the model file ...
Substitution of endo leads >= 2: added 12 auxiliary variables and equations.
Substitution of endo lags >= 2: added 22 auxiliary variables and equations.
Found 267 equation(s).
Evaluating expressions...done
Computing static model derivatives:
- order 1
Computing dynamic model derivatives:
- order 1
- order 2
Processing outputs ...done
Preprocessing completed.
Starting MATLAB/Octave computing.

Total computing time : 0h00m00s
??? Input argument "oo" is undefined.

Error in ==> resol at 96
if isfield(oo,'dr');

Error in ==> stoch_simul_MMB at 50
[oo_.dr, info] = resol(oo_.steady_state,0); % solve

Error in ==> MMB at 544
modelbase=stoch_simul_MMB(modelbase); %
solve model

[HoutanBastani]

This is not a bug in Dynare but rather in stoch_simul_MMB.m. You can see from the error message you posted that it calls resol.m with:

Code: Select all

[oo_.dr, info] = resol(oo_.steady_state,0);

whereas, the arguments to resol changed on 7 Sept 2011 (a while after the 4.2.0 release, so it would only be included in 4.3.X). The new arguments to resol are:

Code: Select all

[dr,info,M,options,oo] = resol(check_flag,M,options,oo)

Hence, to solve this problem, you should be able to replace the line in stoch_simul_MMB.m with

Code: Select all

[~, info, M_, options_, oo_] = resol(0, M_, options_, oo_);

Finally, as posted here http://macromodelbase.com/download, the macro database only works with Dynare 4.2 so you may encounter other problems even after fixing this....

[vvd]

Thanks for getting back to me. It is good to know where the origins of the problem.
As a side note, your suggestion did solve the problem with stoch_simul_MMB.m, but indeed there are other incompatible pieces of code left elsewhere.