stepan
/
dynare
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Allow k order approximation in SIP filter.

rm-particles^2
Stéphane Adjemian (Charybdis) 2019-12-20 11:21:13 +01:00
parent a0fb5c7348
commit 13d79cb438
1 changed files with 17 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
src/sequential_importance_particle_filter.m
View File

@ -1,4 +1,4 @@
function [LIK,lik] = sequential_importance_particle_filter(ReducedForm,Y,start,ParticleOptions,ThreadsOptions)
function [LIK,lik] = sequential_importance_particle_filter(ReducedForm,Y,start,ParticleOptions,ThreadsOptions, DynareOptions, Model)
% Evaluates the likelihood of a nonlinear model with a particle filter (optionally with resampling).
@ -49,14 +49,18 @@ if isempty(init_flag)
init_flag = 1;
end
% Set local state space model (first order approximation).
ghx = ReducedForm.ghx;
ghu = ReducedForm.ghu;
if ReducedForm.use_k_order_solver
dr = ReducedForm.dr;
else
% Set local state space model (first order approximation).
ghx = ReducedForm.ghx;
ghu = ReducedForm.ghu;
% Set local state space model (second order approximation).
ghxx = ReducedForm.ghxx;
ghuu = ReducedForm.ghuu;
ghxu = ReducedForm.ghxu;
% Set local state space model (second order approximation).
ghxx = ReducedForm.ghxx;
ghuu = ReducedForm.ghuu;
ghxu = ReducedForm.ghxu;
end
% Get covariance matrices.
Q = ReducedForm.Q; % Covariance matrix of the structural innovations.
@ -101,7 +105,11 @@ for t=1:sample_size
yhat_ = bsxfun(@minus,StateVectors_,state_variables_steady_state);
[tmp, tmp_] = local_state_space_iteration_2(yhat,epsilon,ghx,ghu,constant,ghxx,ghuu,ghxu,yhat_,steadystate,ThreadsOptions.local_state_space_iteration_2);
else
tmp = local_state_space_iteration_2(yhat,epsilon,ghx,ghu,constant,ghxx,ghuu,ghxu,ThreadsOptions.local_state_space_iteration_2);
if ReducedForm.use_k_order_solver
tmp = local_state_space_iteration_k(yhat, epsilon, dr, Model, DynareOptions);
else
tmp = local_state_space_iteration_2(yhat,epsilon,ghx,ghu,constant,ghxx,ghuu,ghxu,ThreadsOptions.local_state_space_iteration_2);
end
end
%PredictedObservedMean = tmp(mf1,:)*transpose(weights);
PredictionError = bsxfun(@minus,Y(:,t),tmp(mf1,:));