1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
|
function osr_res = osr1(i_params,i_var,weights)
% Compute the Optimal Simple Rules
% INPUTS
% i_params vector index of optimizing parameters in M_.params
% i_var vector variables indices in declaration order
% weights vector weights in the OSRs
%
% OUTPUTS
% osr_res: [structure] results structure containing:
% - objective_function [scalar double] value of the objective
% function at the optimum
% - optim_params [structure] parameter values at the optimum
%
% Algorithm:
%
% Uses Newton-type optimizer csminwel to directly solve quadratic
% osr-problem
%
% Copyright (C) 2005-2017 Dynare Team
%
% This file is part of Dynare.
%
% Dynare is free software: you can redistribute it and/or modify
% it under the terms of the GNU General Public License as published by
% the Free Software Foundation, either version 3 of the License, or
% (at your option) any later version.
%
% Dynare is distributed in the hope that it will be useful,
% but WITHOUT ANY WARRANTY; without even the implied warranty of
% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
% GNU General Public License for more details.
%
% You should have received a copy of the GNU General Public License
% along with Dynare. If not, see <http://www.gnu.org/licenses/>.
global M_ oo_ options_ it_
klen = M_.maximum_lag + M_.maximum_lead + 1;
iyv = M_.lead_lag_incidence';
iyv = iyv(:);
iyr0 = find(iyv) ;
it_ = M_.maximum_lag + 1 ;
osr_res.error_indicator = 1; %initialize indicator
if M_.exo_nbr == 0
oo_.exo_steady_state = [] ;
end
if ~ M_.lead_lag_incidence(M_.maximum_lag+1,:) > 0
error ('OSR: Error in model specification: some variables don''t appear as current') ;
end
if M_.maximum_lead == 0
error ('OSR: Backward or static model: no point in using OSR') ;
end
if any(any(isinf(weights)))
error ('OSR: At least one of the optim_weights is infinite.') ;
end
if any(isnan(M_.params(i_params)))
error ('OSR: At least one of the initial parameter values for osr_params is NaN') ;
end
%restore backward compatibility with maxit and tolf
if isfield(options_.osr,'maxit') || isfield(options_.osr,'tolf')
warning('OSR: The use of maxit and tolf is now deprecated. Please use the optim-option instead.')
if options_.osr.opt_algo~=4
error ('OSR: The maxit and tolf options are not supported when not using the default opt_algo=4. Use the optim-option instead.') ;
else
if isfield(options_.osr,'maxit')
if ~isempty(options_.optim_opt)
options_.optim_opt=[options_.optim_opt,','];
end
options_.optim_opt=[options_.optim_opt,'''MaxIter'',',num2str(options_.osr.maxit),''];
end
if isfield(options_.osr,'tolf')
if ~isempty(options_.optim_opt)
options_.optim_opt=[options_.optim_opt,','];
end
options_.optim_opt=[options_.optim_opt,'''TolFun'',',num2str(options_.osr.tolf),''];
end
end
end
exe =zeros(M_.exo_nbr,1);
oo_.dr = set_state_space(oo_.dr,M_,options_);
np = size(i_params,1);
t0 = M_.params(i_params);
inv_order_var = oo_.dr.inv_order_var;
%extract unique entries of covariance
i_var=unique(i_var);
%% do initial checks
[loss,info,exit_flag,vx]=osr_obj(t0,i_params,inv_order_var(i_var),weights(i_var,i_var));
if info~=0
print_info(info, options_.noprint, options_);
else
if ~options_.noprint
fprintf('\nOSR: Initial value of the objective function: %g \n\n',loss);
end
end
if ~options_.noprint && isinf(loss)
fprintf('\nOSR: The initial value of the objective function is infinite.\n');
fprintf('\nOSR: Check whether the unconditional variance of a target variable is infinite\n');
fprintf('\nOSR: due to the presence of a unit root.\n');
error('OSR: Initial likelihood is infinite')
end
if isequal(options_.osr.opt_algo,5)
error('OSR: OSR does not support opt_algo=5.')
elseif isequal(options_.osr.opt_algo,6)
error('OSR: OSR does not support opt_algo=6.')
elseif isequal(options_.osr.opt_algo,10)
error('OSR: OSR does not support opt_algo=10.')
elseif isequal(options_.osr.opt_algo,11)
error('OSR: OSR does not support opt_algo=11.')
else
if ~isempty(M_.osr.param_bounds) && ~(ismember(options_.osr.opt_algo,[1,2,5,9]) || ischar(options_.osr.opt_algo))
error('OSR: OSR with bounds on parameters requires a constrained optimizer, i.e. 1,2,5, or 9.')
end
%%do actual optimization
[p, f, exitflag] = dynare_minimize_objective(str2func('osr_obj'),t0,options_.osr.opt_algo,options_,M_.osr.param_bounds,cellstr(M_.param_names(i_params,:)),[],[], i_params,...
inv_order_var(i_var),weights(i_var,i_var));
end
osr_res.objective_function = f;
M_.params(i_params)=p; %make sure optimal parameters are set (and not the last draw used in csminwel)
for i=1:length(i_params)
osr_res.optim_params.(deblank(M_.param_names(i_params(i),:))) = p(i);
end
if ~options_.noprint
skipline()
disp('OPTIMAL VALUE OF THE PARAMETERS:')
skipline()
for i=1:np
disp(sprintf('%16s %16.6g\n',M_.param_names(i_params(i),:),p(i)))
end
disp(sprintf('Objective function : %16.6g\n',f));
skipline()
end
[oo_.dr,info,M_,options_,oo_] = resol(0,M_,options_,oo_);
if ~info
osr_res.error_indicator=0;
end
|
