Estimation with dynare 4.5.7

Dear Everyone

I am using dynare 4.5.7 to estimate my model as the attached file File.zip (15.9 KB)

I wondering that this file can not use the mode-compute method number 9 (mode_compute=9). Indeed, only two interactions are runned, then finished

Blockquote
Initial value of the log posterior (or likelihood): 353.0099
n=22: (6,13)-CMA-ES(w=[40 25 17 10 6 2]%, mu_eff=3.7) on function dsge_likelihood
Warning: Non-finite fitness range
In cmaes (line 974)
In dynare_minimize_objective (line 360)
In dynare_estimation_1 (line 220)
In dynare_estimation (line 105)
In singaporeversion2 (line 341)
In dynare (line 235)
In MainSingapore (line 11)
Iterat, #Fevals: Function Value (median,worst) |Axis Ratio|idx:Min SD idx:Max SD
1 , 14 : 2.9599924933286e+03 +(Inf,Inf) | 1.34e+04 | 1:2.4e-02 22:3.2e+02
Warning: Non-finite fitness range
In cmaes (line 974)
In dynare_minimize_objective (line 360)
In dynare_estimation_1 (line 220)
In dynare_estimation (line 105)
In singaporeversion2 (line 341)
In dynare (line 235)
In MainSingapore (line 11)
2 , 27 : Inf +(NaN,NaN) | 1.34e+04 | 1:2.3e-02 17:3.1e+02
#Fevals: f(returned x) | bestever.f | stopflag (saved to variablescmaes.mat)
28: Inf | -3.53009901654e+02 | equalfunvals
mean solution: +9.1e-01 +3.9e+00 -4.1e-02 +2.7e-01 +3.3e+00 +1.9e+00 +1.8e+00 +4.2e+00 +5.1e-01 +1.1e+00 +1.1e+00 +7.5e-01 +7.8e-01 +5.1e-01 +2.2e+02 +2.2e+02 +4.2e+02 +2.2e+02 -9.3e+01 +2.6e+01 -1.6e+02 -2.3e+02
std deviation: 2.3e-02 5.8e-01 2.6e-02 3.8e-02 2.9e+00 9.0e-02 1.1e+00 4.7e+00 1.8e-01 1.2e-01 1.8e-01 1.8e-01 1.8e-01 1.8e-01 3.0e+02 3.0e+02 3.1e+02 3.0e+02 3.0e+02 3.0e+02 3.0e+02 3.0e+02
use plotcmaesdat.m for plotting the output at any time (option LogModulo must not be zero)
Final value of minus the log posterior (or likelihood):-353.009902
POSTERIOR KERNEL OPTIMIZATION PROBLEM!
(minus) the hessian matrix at the “mode” is not positive definite!
=> posterior variance of the estimated parameters are not positive.
You should try to change the initial values of the parameters using
the estimated_params_init block, or use another optimization routine.

I use other method, such as mode_compute=7, then it runs

Thank you so much indeed

Best wishes

Your model causes some problem with the default settings of the optimizer. I will have to investigate this more deeply.