Page 1 of 1

Problem with conditional_forecast

PostPosted: Fri Feb 19, 2010 9:26 pm
by doctorwes
My .mod file contains the following:

Code: Select all
conditional_forecast_paths;
var x;
periods 1:1, 2:3, 4:4;
values 1.5, -0.5, 0.;
var r;
periods 1:4;
values -2.;
end;

conditional_forecast(parameter_set = posterior_mean, controlled_varexo = (d, m));

plot_conditional_forecast x r;


I'm getting the following error:

Code: Select all
??? Undefined variable or function iF might refer to the function if.

Error in ==> missing_DiffuseKalmanSmoother3 at 295
        ri_d = Z'*iF(:,:,t)*v(:,t);

Error in ==> DsgeSmoother at 272
            [alphahat,etahat,ahat,P,aK,PK,d,decomp] = missing_DiffuseKalmanSmoother3(T,R,Q,
            ...

Error in ==> imcforecast at 130
[atT,innov,measurement_error,filtered_state_vector,ys,trend_coeff] =
DsgeSmoother(xparam,gend,data,data_index,number_of_observations);

Error in ==> iccbe at 288
imcforecast(constrained_paths_, constrained_vars_, options_cond_fcst_);

Error in ==> dynare at 132
evalin('base',fname) ;


Am I invoking this command incorrectly? The estimation itself runs successfully.

Re: Problem with conditional_forecast

PostPosted: Sat Feb 20, 2010 12:28 am
by StephaneAdjemian
Hi,

I think that function missing_DiffuseKalmanSmoother3 is broken. But I don't understand why dynare goes there. Do you have NaNs in your dataset ?

Best,
Stéphane.

doctorwes wrote:My .mod file contains the following:

Code: Select all
conditional_forecast_paths;
var x;
periods 1:1, 2:3, 4:4;
values 1.5, -0.5, 0.;
var r;
periods 1:4;
values -2.;
end;

conditional_forecast(parameter_set = posterior_mean, controlled_varexo = (d, m));

plot_conditional_forecast x r;


I'm getting the following error:

Code: Select all
??? Undefined variable or function iF might refer to the function if.

Error in ==> missing_DiffuseKalmanSmoother3 at 295
        ri_d = Z'*iF(:,:,t)*v(:,t);

Error in ==> DsgeSmoother at 272
            [alphahat,etahat,ahat,P,aK,PK,d,decomp] = missing_DiffuseKalmanSmoother3(T,R,Q,
            ...

Error in ==> imcforecast at 130
[atT,innov,measurement_error,filtered_state_vector,ys,trend_coeff] =
DsgeSmoother(xparam,gend,data,data_index,number_of_observations);

Error in ==> iccbe at 288
imcforecast(constrained_paths_, constrained_vars_, options_cond_fcst_);

Error in ==> dynare at 132
evalin('base',fname) ;


Am I invoking this command incorrectly? The estimation itself runs successfully.