Suppress optimization output

Estimation ML/Bayesian estimation
1

Dear all,

I am using Dynare for Monte Carlo experiment in estimating a DSGE model. I use ML estimation and would like to suppress the optimization output to increase the speed.
My code looks like that:

options_.console_mode=1;
options_.nograph=1;
options_.silent_optimizer=1;
options_.cova_compute=0;
options_.plot_priors=0;
options_.mh_replic= 0;
options_.noprint=1;
options_.lik_init=1;

for ii=1:10
estimation(datafile = '../../data_simulation/simulated_data_files/calibration1_n1.csv', mode_compute=5, nograph, nodisplay);
end

Is it also possible to suppress that:

Final value of minus the log posterior (or likelihood):178.744820 

RESULTS FROM MAXIMUM LIKELIHOOD ESTIMATION
parameters
         Estimate    s.d. t-stat

omeg_f     0.3087     NaN     NaN 
gam        0.1099     NaN     NaN 
bet_f      0.2968     NaN     NaN 
bet_r      0.0678     NaN     NaN 
rho        0.3426     NaN     NaN 
gam_infl   1.7340     NaN     NaN 
gam_y      0.2748     NaN     NaN 
rho_y      0.5172     NaN     NaN 
rho_r      0.8214     NaN     NaN 

standard deviation of shocks
         Estimate    s.d. t-stat

u_y        0.2935     NaN     NaN 
u_r        0.2666     NaN     NaN
2

No, that is currently not feasible. But it is also not recommended to loop over an estimation command. There are a lot more efficient ways. What are you varying in your Monte Carlo study?

3

I am varying only parameters in the NKPC. What do you mean by more efficient ways - still using dynare or standalone code?

4

With your call to the estimation-command you are running dynare_estimation_1.m, which is a very complex function. For Monte Carlo studies it is not recommended to run this whole function repeatedly, but rather only the parts you need. So it will still use Dynare, but will require some adapted Matlab programming.