Estimating the correlation between shocks

As Dynare 4 is now the main version, this forum is now closed. Please, post
on the Dynare Help forum
Forum rules
As Dynare 4 is the current Dynare version, there forum is now close. Post new question in the Dynare Help Forum.

Estimating the correlation between shocks

Postby almira » Tue Feb 12, 2008 3:34 pm

Dear all,
I got some troubles using Dynare 4 by estimating the correlation between shocks. So in the "shocks" section I just assumed that the correlation coefficient between shocks is equal to 1:

shocks;
var e_a; stderr 0.5;
var e_a, e_x = 0.5*0.5;
var e_ph; stderr 0.5;
var e_ph, e_x = 0.5*0.5;
var e_t; stderr 0.5;
var e_t, e_x = 0.5*0.5;
var e_th; stderr 0.5;
var e_th, e_x = 0.5*0.5;
var e_x; stderr 0.5;
end;

In the "estimated_params" section I specified correlation coefficients to be estimated using MLE as follows:
estimated_params;
stderr e_a,0.5,,;
corr e_a, e_x,0.123,-1,1;
stderr e_ph,0.5,,;
corr e_ph, e_x,0.152,-1,1;
stderr e_t,0.5,,;
corr e_t, e_x,-0.845,-1,1;
stderr e_th,0.5,,;
corr e_th, e_x,-0.427,-1,1;
stderr e_x,0.5,,;
end;

After running the .mod-file I got estimation results for standard deviation and correlation coefficients of specified shocks
AND the following error message:

??? Undefined command/function 'Sigma_e_'.

Error in ==> set_all_parameters at 42
Sigma_e(k1,k2) = xparam1(i+offset)*sqrt(Sigma_e_(k1,k1)*Sigma_e_(k2,k2));

Error in ==> DsgeSmoother at 35
set_all_parameters(xparam1);

Error in ==> dynare_estimation at 887
[atT,innov,measurement_error,filtered_state_vector,ys,trend_coeff] = DsgeSmoother(xparam1,gend,data);

Error in ==> test12_2 at 284
dynare_estimation(var_list_);

Error in ==> dynare at 79
evalin('base',fname) ;

Until now I have not found out, why this error message shows up.
I would be glad if somebody knows about this error and could explain me how to deal with that.
Thank you!
Almira
almira
 
Posts: 2
Joined: Mon Feb 11, 2008 1:43 pm
Location: University of Bonn

Postby MichelJuillard » Sat Feb 16, 2008 7:34 pm

We just corrected bugs relative to estimation of models with measurement error. Please, grab the last snapshot.
Thanks for reporting the problem

Best

Michel
MichelJuillard
 
Posts: 680
Joined: Thu Nov 18, 2004 10:51 am


Return to Dynare version 4

Who is online

Users browsing this forum: No registered users and 1 guest