Kalman filter

Dynare help (legacy posts)
1

Hi,

I have received an error message as shown below when I run estimation. Could anybody tell me what the problem is and how to solve it? Thanks in advance!

??? Attempted to access H(2); index out of bounds because numel(H)=1.

Error in ==> univariate_kalman_filter at 170
Fi = z(i,:)*PZ + H(d_index(i));

Error in ==> dsge_likelihood at 686
[LIK, lik] =
univariate_kalman_filter(DynareDataset.missing.aindex,DynareDataset.missing.number_of_observations,DynareDataset.missing.no_more_missing_observations,Y,diffuse_periods+1,size(Y,2),
.
Error in ==> initial_estimation_checks at 45
[fval,junk1,junk2,a,b,c,d] =
feval(objective_function,xparam1,DynareDataset,DynareOptions,Model,EstimatedParameters,BayesInfo,DynareResults);

Error in ==> dynare_estimation_1 at 133
oo_ =
initial_estimation_checks(objective_function,xparam1,dataset_,M_,estim_params_,options_,bayestopt_,oo_);

Error in ==> dynare_estimation at 70
dynare_estimation_1(var_list,dname);

2

Are you using the most recent Dynare version? If yes, please post the mod- and the datafile.

3

Hi, I am having the same problem. Unfortunately, I am not authorized to send the mod file neither the data file. The mod file works with loops and I am trying to estimate two different models separatelly. Basically, the loop changes the number of agents in the economy, for example, let be ‘i’ the number of a certain type of agent, and each agent has different consumption that we denote c_i. Thus, if i = 10, we have a time series of c_1, c_2, …, c_10 in the model, if i = 30, we have c_1, c_2, … c_30. When I set i = 10, the mode file works well, however, when I set i = 30, the error message comes. Only for reason of testing, I increase the ‘i’ and the problem show up for all i > 23. The ‘c_i’ are observable variable among others.

I really appreciate any help, given that limited information, without the code.

Best,
rso.

4

Have you tried the most recent snapshot or 4.3.2?