1 Blanchard Kahn conditions are not satisfied: indeterminacy

Estimation
1

Dear Professor, I have the following problems when using Bayesian estimation. I hope you can help me.
错误使用 print_info (line 45)
Blanchard Kahn conditions are not satisfied: indeterminacy

出错 print_info (line 45)
error([‘Blanchard Kahn conditions are not satisfied:’ …

出错 initial_estimation_checks (line 175)
print_info(info, DynareOptions.noprint, DynareOptions)

出错 dynare_estimation_1 (line 165)
oo_ =
initial_estimation_checks(objective_function,xparam1,dataset_,dataset_info,M_,estim_params_,options_,bayestopt_,bounds,oo_);

出错 dynare_estimation (line 105)
dynare_estimation_1(var_list,dname);

出错 bys2 (line 376)
oo_recursive_=dynare_estimation(var_list_);

出错 dynare (line 235)
evalin(‘base’,fname) ;

2

Use

estimated_params_init(use_calibration);
end;

It will reveal that you have stochastic singularity. Search the forum on this.

3

Dear Professor, after I modified the code, the problem you mentioned appeared. I searched for relevant solutions on the forum, but the problem still could not be solved. Therefore, I would like to ask you how to solve this problem.
initial_estimation_checks:: The forecast error variance in the multivariate Kalman filter became singular.
This is my code.bys2.mod (3.6 KB) sysj2.xlsx (15.3 KB)

4

Stochastic singularity happens when your observables imply an exact linear combination. For example, in your model

chi=mg-mc;

so you cannot observe all three without measurement error.

5

Dear Professor, I have added measurement errors to all the observed variables in my code, but why does such a problem exist?I hope you can help me solve this problem.Here’s my code and my data.
bys2.mod (3.7 KB) sysj2.xlsx (15.3 KB)

6

You defined measurement error, but set its variance to the default of 0.

7
  • Thanks to the professor’s help, my problem was solved