Hi,

I am trying to estimate a model using 2 observables Y,C and also 2 parameters using the below block:

estimated_params;

kappa_p, Gamma_pdf,40.5,0.5;

theta, Gamma_pdf,18,0.5;

end;

varobs Y,C;

estimation(datafile=Euro,mh_nblocks=3,mh_replic=5000,mh_jscale=3,mh_init_scale=12) C,Y;

BUT i receive the message:

"Error using NaN

NaN and Inf not allowed.

Error in : (line 79)

C.time = NaN(m,2);

Error in dseries (line 137)

ts.dates = init:init+(nobs(ts)-1);

Error in makedataset (line 121)

DynareDataset = dseries(datafile);

Error in dynare_estimation_init (line 538)

[dataset_, dataset_info, newdatainterfaceflag] = makedataset(options_,

options_.dsge_var*options_.dsge_varlag, gsa_flag);

Error in dynare_estimation_1 (line 116)

dynare_estimation_init(var_list_, dname, [], M_, options_, oo_, estim_params_, bayestopt_);

Error in dynare_estimation (line 105)

dynare_estimation_1(var_list,dname);

Error in Baeysian_DSGE_19 (line 510)

oo_recursive_=dynare_estimation(var_list_);

Error in dynare (line 235)

evalin(‘base’,fname) ; "

the model is solved properly with the stoch_simul command but not with Bayesian estimation.

Any help?

Mike

There is an issue with your data-file, but without seeing the file, it is impossible to tell.

@jpfeifer attached you may find the data file (it contains quarterly data for Y and C) and it is stored in C:\dynare\4.5.7\matlab which is also the path with the rest mod file.

Attached you may find also the mod file.

Bayesian test.mod (6.7 KB)

Euro.xls (22 KB)

When I run your code, Dynare complains about the steady state not being found. Is there a steady state file missing?

In any case, the variables in your data file contain a trend. That will not work.

@jpfeifer, after reading your paper i tried to include 2 simple observables equations in the model in order to estimate. However while the model runs and produces results i have troubles with the estimation part.

I receive the below message:

Error using dynare_estimation_init (line 352)

Estimation: the ‘estimated_params’ block is mandatory (unless you are running a smoother)

Error in dynare_estimation_1 (line 116)

dynare_estimation_init(var_list_, dname, [], M_, options_, oo_, estim_params_, bayestopt_);

Error in dynare_estimation (line 105)

dynare_estimation_1(var_list,dname);

Error in Baeysian_test (line 473)

oo_recursive_=dynare_estimation(var_list_);

Error in dynare (line 235)

evalin(‘base’,fname) ;

What exactly should i do with the estimated_params command? Do i need to add some parameters? I thought that when i input observables it is not neccesary to add them

Attached you may find the code and the file use.

Baeysian_test.mod (6.7 KB) Simuldata.xls (20.5 KB)

Regards

Michael

If you want to do Bayesian estimation, you need to tell Dynare which parameters are supposed to be estimated. Otherwise, there is nothing to do.