Dynare forums

Hi, when i run the simulation, the BK condition is no problem, but when I add the estimation, Dynare says the BK condition is not satisfied, how come this is a problem? Why would BK condition satisfied and not satisfied in the same time?

Configuring Dynare ...
[mex] Generalized QZ.
[mex] Sylvester equation solution.
[mex] Kronecker products.
[mex] Sparse kronecker products.
[mex] Local state space iteration (second order).
[mex] Bytecode evaluation.
[mex] k-order perturbation solver.
[mex] k-order solution simulation.
[mex] Quasi Monte-Carlo sequence (Sobol).
[mex] Markov Switching SBVAR.

Starting Dynare (version 4.3.1).
Starting preprocessing of the model file ...
Found 18 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.


EIGENVALUES:
Modulus Real Imaginary

0 -0 0
0.5631 0.5631 0
0.9 0.9 0
0.9 0.9 0
1 1 0
1 1 0
1.005 1.003 0.0588
1.005 1.003 -0.0588
1.794 1.794 0
Inf Inf 0


There are 4 eigenvalue(s) larger than 1 in modulus
for 4 forward-looking variable(s)

The rank condition is verified.


MODEL SUMMARY

Number of variables: 18
Number of stochastic shocks: 2
Number of state variables: 6
Number of jumpers: 4
Number of static variables: 8


MATRIX OF COVARIANCE OF EXOGENOUS SHOCKS

Variables eps_a_os eps_a
eps_a_os 4.000000 0.000000
eps_a 0.000000 1.000000

POLICY AND TRANSITION FUNCTIONS
ygap ppi_h r ppi e q p_h cpi_level
a_os(-1) 0.550790 0.349210 -0.090000 0.209526 0 -0.209526 0.349210 0.209526
a(-1) -0.550790 -0.349210 0 -0.209526 0 0.209526 -0.349210 -0.209526
s(-1) 0.563109 0.436891 0 0.262135 0 0.337865 0.436891 0.262135
e(-1) -0.563109 -0.436891 0 -0.662135 0 -0.337865 -0.436891 -0.662135
p_h(-1) 0 0 0 0 0 0 1.000000 0
cpi_level(-1) 0 0 0 0 0 0 0 1.000000
eps_a_os 0.140758 0.089242 -0.100000 0.053545 0 -0.053545 0.089242 0.053545
eps_a -0.611989 -0.388011 0 -0.232806 0 0.232806 -0.388011 -0.232806


MOMENTS OF SIMULATED VARIABLES

VARIABLE MEAN STD. DEV. VARIANCE SKEWNESS KURTOSIS
ygap 0.049970 0.757897 0.574408 -0.162758 -0.244163
ppi_h -0.003343 0.484963 0.235189 -0.228369 -0.184871
r 0.234266 0.361072 0.130373 0.018553 -0.153323
ppi -0.002006 0.290978 0.084668 -0.228369 -0.184871
e 0.000000 0.000000 0.000000 0.064579 -0.563350
q -0.261102 0.936058 0.876205 0.128145 -1.065849
p_h 0.435170 1.560097 2.433904 -0.128145 -1.065849
cpi_level 0.261102 0.936058 0.876205 -0.128145 -1.065849


CORRELATION OF SIMULATED VARIABLES

VARIABLE ygap ppi_h r ppi e q p_h cpi_level
ygap 1.0000 0.9665 -0.4969 0.9665 -0.0614 -0.3950 0.3950 0.3950
ppi_h 0.9665 1.0000 -0.4406 1.0000 -0.0286 -0.1461 0.1461 0.1461
r -0.4969 -0.4406 1.0000 -0.4406 0.7297 0.3385 -0.3385 -0.3385
ppi 0.9665 1.0000 -0.4406 1.0000 -0.0286 -0.1461 0.1461 0.1461
e -0.0614 -0.0286 0.7297 -0.0286 1.0000 0.1342 -0.1342 -0.1342
q -0.3950 -0.1461 0.3385 -0.1461 0.1342 1.0000 -1.0000 -1.0000
p_h 0.3950 0.1461 -0.3385 0.1461 -0.1342 -1.0000 1.0000 1.0000
cpi_level 0.3950 0.1461 -0.3385 0.1461 -0.1342 -1.0000 1.0000 1.0000


AUTOCORRELATION OF SIMULATED VARIABLES

VARIABLE 1 2 3 4 5
ygap 0.4395 0.2121 0.0641 -0.0065 0.0641
ppi_h 0.4058 0.1723 0.0252 -0.0368 0.0459
r 0.8267 0.6610 0.5010 0.4347 0.3583
ppi 0.4058 0.1723 0.0252 -0.0368 0.0459
e 0.7908 0.6166 0.4381 0.3424 0.2352
q 0.9443 0.8465 0.7292 0.6123 0.5092
p_h 0.9443 0.8465 0.7292 0.6123 0.5092
cpi_level 0.9443 0.8465 0.7292 0.6123 0.5092
Loading 200 observations from data.mat

Error in computing likelihood for initial parameter values
Error using print_info (line 40)
Blanchard Kahn conditions are not satisfied: no stable equilibrium

Error in initial_estimation_checks (line 69)
print_info(info, DynareOptions.noprint)

Error in dynare_estimation_1 (line 147)
oo_ =
initial_estimation_checks(objective_function,xparam1,dataset_,M_,estim_params_,options_,bayestopt_,oo_);

Error in dynare_estimation (line 70)
dynare_estimation_1(var_list,dname);

Error in gali_esti (line 229)
dynare_estimation(var_list_);

Error in dynare (line 120)
evalin('base',fname) ;

It's probably that your initial values of parameters for estimation differ from values used for simulation. There is alternative posibility that you have problem with numerical steady calculation (but probability of such case is small). The mod-file is needed for detailed check.

Were you able to solve your problem with BK in the estimation? I'm running into the same problem. The worst is that I set the priors at exactly the same value that was calibrated in the simulation, and get B&K in the estimation, but not in the simulation. When I run the command model_diagnostics, it tells me
Error using svd
Input to SVD must not contain NaN or Inf.

Error in rank (line 15)
s = svd(A);

Error in model_diagnostics (line 104)
rank_jacob = rank(jacob);
which doesn't help.

If I do the same thing at dynare 4.2.5, the command returns nothing.

If anyone knows what is really wrong, please let me know.

You have two unit roots. Those are treated differently in estimation and simulation. Hence, the error. See the manual regarding lik_init and qz_crit.