DsgeVarLikelihood of Dynare 4.0.0 vs. Inestable Version

[Aldo]

I am using codes written by Tim Robinson to compute forecast error (RMSE), in Dynare 4.0.0, especifically the DsgeVarLikelihood function:

[fval, cost_flag, info, PHI_STACKED, SIGMA, iXX] = DsgeVarLikelihood(xparam1, gend)

where:

xparam1 : value of vector of estimated parameters
gend : Number of observations in the sample

I would like to use the unstable version of Dynare because it is faster than Dynare 4.0.0, but in this last I get this error:

Error using dsge_var_likelihood (line 80)
Not enough input arguments.

Could you please help me to use the dsge_var_likelihood function and define inputs in the inestable version?

Also, I have another question, when I invoke options_.nobs in inestable version I get a different value of the 4.0.0 version.

Thanks in advance

Aldo

[jpfeifer]

In the unstable version, the syntax is

Code: Select all

[fval,info,exit_flag,grad,hess,SteadyState,trend_coeff,PHI_tilde,SIGMA_u_tilde,iXX,prior] =dsge_var_likelihood(xparam1,dataset_,dataset_info,options_,M_,estim_params_,bayestopt_,bounds,oo_)

where most of the inputs come from the call to

Code: Select all

[dataset_, dataset_info, xparam1, hh, M_, options_, oo_, estim_params_, bayestopt_, bounds] = ...
    dynare_estimation_init(var_list_, dname, [], M_, options_, oo_, estim_params_, bayestopt_);

Regarding your second question: what exactly do you mean? I would need more details

[Aldo]

Thanks jpfeifer,

1. About the last question, I mean when I run my .mod I get options_.nobs = 32 in unstable version, which is size of my data,
but in 4.4.3 and previous versions I get options_.nobs = 28.

2. By other side, when I try to use dsge_var_likelihood function running this:

Code: Select all

   dynare mymod   

   options_.first_obs = 5;
   options_.nobs = NaN;
   [dataset_, dataset_info, xparam1, hh, M_, options_, oo_, estim_params_, bayestopt_, bounds] = ...
   dynare_estimation_init(var_list_, [], [], M_, options_, oo_, estim_params_, bayestopt_);


   [fval,info,exit_flag,grad,hess,SteadyState,trend_coeff,PHI_tilde,SIGMA_u_tilde,iXX,prior] = ...
    dsge_var_likelihood(xparam1,dataset_,dataset_info,options_,M_,estim_params_,bayestopt_,bounds,oo_);

I get size(PHI_tilde,1) = 20. But It should be 10 (=5x2) because I have five observable variables and I set

Code: Select all

options_.varlag = 2;


3. What would be the rigth form to loop over

Code: Select all

dsge_prior_weight

option and over sample size to get PHI_tilde and SIGMA_u_tilde. I loop over dsge_prior_weight in a .mod and I loop over sample size in a .m file. But I see when I loop in a .mod (over dsge_prior_weight) dynare estimates twice, it makes slow this method, because in each iteration I run my .mod and after I run dynare_estimation(var_list_). I attach my files.

[jpfeifer]

1- That is a matter of the changed internal treatment. In 4.4.3 nobs deducted the options_.first_obs. But that should not affect the results.
2. You are setting the wrong option for the lag number. It should be

Code: Select all

options_.dsge_varlag

Thus, you are still at the default of 4.
3. You should be able to change the prior weight within a loop via

Code: Select all

set_param_value('dsge_prior_weight',1)