Identification does not work in 4.2.1

Dynare help (legacy posts)
1

Attempts to call the indentification routine in Dynare produce the following error message:

??? Error: File: dynare_identification.m Line: 759 Column: 11
Expression or statement is incorrect–possibly unbalanced (, {, or .

When I look at the code listed in the error I see the following,

    ~,is]=sort(idelre.cond);

This does not look like correct Malab code. Is this supposed to be

    [ss,is]=sort(idelre.cond);

(just taking a wild guess based on a very quick look at earlier versions of the code, but without understanding the routine at all)?

2

The tilde operator has been introduced in Matlab 2009b, i.e. version 7.9. It is used to suppress unused function output. I will not run on earlier versions. Apparently, there is a check for the Matlab version missing.
A way to work around this problem is to do as you suggested and replace it with any new variable name that is not subsequently used.

3

Thank you very much for the reply. I will try your suggestion.

4

Now I have found a new problem (following my repair as suggested above to fix the lack of support in R2008b for ~,is] = etc.)
Here is the error message:

==== Identification analysis ====

Testing prior mean
??? Error using ==> feval
Error: File: Model_14F_cal_S_stationary_NUR_withL_id_params_derivs.m Line:
1950 Column: 221011
Nesting of {, , and ( cannot exceed a depth of 32.

Error in ==> getH at 34
df = feval([M_.fname,‘params_derivs’],yy0, oo.exo_steady_state’, …

Error in ==> getJJ at 39
[H, dA, dOm, dYss, gp] = getH(A, B, M_,oo_,kronflag,indx,indexo);

Error in ==> dynare_identification at 260
[JJ, H, gam, gp] = getJJ(A, B,
M_,oo0,options_,0,indx,indexo,bayestopt_.mf2,nlags,useautocorr);

Error in ==> Model_14F_cal_S_stationary_NUR_withL_id at 397
dynare_identification(options_ident);

Error in ==> dynare at 132
evalin(‘base’,fname) ;

Any ideas?
Thank you.

5

Hi,

Can you post the MOD file which triggers this error, so that we can reproduce the problem and, hopefully, fix it?

Thanks