From 23d49470c80b950f3beace3458ca45dd6b260960 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?S=C3=A9bastien=20Villemot?= <sebastien@dynare.org>
Date: Fri, 4 Apr 2025 15:17:00 +0200
Subject: [PATCH] =?UTF-8?q?Make=20perfect=5Fforesight=5Fcontrolled=5Fpaths?=
=?UTF-8?q?=20compatible=20with=20the=20=E2=80=9Clinear=E2=80=9D=20option?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
---
doc/manual/source/the-model-file.rst | 4 +-
.../perfect_foresight_solver_core.m | 5 +-
matlab/perfect-foresight-models/sim1_linear.m | 41 +++++++++++--
meson.build | 1 +
.../linear_state_space_arma_condpaths.mod | 60 +++++++++++++++++++
5 files changed, 101 insertions(+), 10 deletions(-)
create mode 100644 tests/deterministic_simulations/linear_state_space_arma_condpaths.mod
diff --git a/doc/manual/source/the-model-file.rst b/doc/manual/source/the-model-file.rst
index c47208e9f0..86c0533844 100644
--- a/doc/manual/source/the-model-file.rst
+++ b/doc/manual/source/the-model-file.rst
@@ -4672,8 +4672,8 @@ that have been left free.
The ``perfect_foresight_controlled_paths`` block requires that the
:opt:`stack_solve_algo <stack_solve_algo = INTEGER>` option be equal to
either ``0``, ``1``, ``2``, ``3``, ``6`` or ``7``, and is incompatible with
- the :opt:`block`, :opt:`linear` and :opt:`bytecode` options of the
- :bck:`model` block and :comm:`model_options` command.
+ the :opt:`block` and :opt:`bytecode` options of the :bck:`model` block and
+ :comm:`model_options` command.
*Options*
diff --git a/matlab/perfect-foresight-models/perfect_foresight_solver_core.m b/matlab/perfect-foresight-models/perfect_foresight_solver_core.m
index c1977137cc..879961b8e9 100644
--- a/matlab/perfect-foresight-models/perfect_foresight_solver_core.m
+++ b/matlab/perfect-foresight-models/perfect_foresight_solver_core.m
@@ -90,9 +90,6 @@ if ~isempty(controlled_paths_by_period)
if options_.block
error('perfect_foresight_controlled_paths is not available with the block option')
end
- if options_.linear
- error('perfect_foresight_controlled_paths is not available with the linear option')
- end
if M_.maximum_endo_lead == 0 || M_.maximum_endo_lag == 0
error('perfect_foresight_controlled_paths is not available with purely backward, purely forward or static models')
end
@@ -147,7 +144,7 @@ else
switch options_.stack_solve_algo
case {0 2 3}
if options_.linear_approximation
- [y, success, maxerror] = sim1_linear(y, exo_simul, steady_state, exo_steady_state, M_, options_);
+ [y, success, maxerror, exo_simul] = sim1_linear(y, exo_simul, steady_state, exo_steady_state, controlled_paths_by_period, M_, options_);
else
[y, success, maxerror, iter, exo_simul] = sim1(y, exo_simul, steady_state, controlled_paths_by_period, M_, options_);
end
diff --git a/matlab/perfect-foresight-models/sim1_linear.m b/matlab/perfect-foresight-models/sim1_linear.m
index 30c11700cf..6524cf22f2 100644
--- a/matlab/perfect-foresight-models/sim1_linear.m
+++ b/matlab/perfect-foresight-models/sim1_linear.m
@@ -1,12 +1,13 @@
-function [endogenousvariables, success, ERR] = sim1_linear(endogenousvariables, exogenousvariables, steadystate_y, steadystate_x, M_, options_)
+function [endogenousvariables, success, ERR, exogenousvariables] = sim1_linear(endogenousvariables, exogenousvariables, steadystate_y, steadystate_x, controlled_paths_by_period, M_, options_)
% [endogenousvariables, success, ERR] = sim1_linear(endogenousvariables, exogenousvariables, steadystate_y, steadystate_x, M_, options_)
% Solves a linear approximation of a perfect foresight model using sparse matrix.
%
% INPUTS
-% - endogenousvariables [double] N*T array, paths for the endogenous variables (initial guess).
+% - endogenousvariables [double] N*T array, paths for the endogenous variables (initial condition + initial guess + terminal condition).
% - exogenousvariables [double] T*M array, paths for the exogenous variables.
% - steadystate_y [double] N*1 array, steady state for the endogenous variables.
% - steadystate_x [double] M*1 array, steady state for the exogenous variables.
+% - controlled_paths_by_period [struct] data from perfect_foresight_controlled_paths block
% - M_ [struct] contains a description of the model.
% - options_ [struct] contains various options.
%
@@ -14,6 +15,8 @@ function [endogenousvariables, success, ERR] = sim1_linear(endogenousvariables,
% - endogenousvariables [double] N*T array, paths for the endogenous variables (solution of the perfect foresight model).
% - success [logical] Whether a solution was found
% - ERR [double] ∞-norm of the residual
+% - exogenousvariables [double] T*M array, paths for the exogenous variables
+% (may be modified if perfect_foresight_controlled_paths present)
%
% NOTATIONS
% - N is the number of endogenous variables.
@@ -39,7 +42,7 @@ function [endogenousvariables, success, ERR] = sim1_linear(endogenousvariables,
% to center the variables around the deterministic steady state to solve the
% perfect foresight model.
-% Copyright © 2015-2024 Dynare Team
+% Copyright © 2015-2025 Dynare Team
%
% This file is part of Dynare.
%
@@ -67,6 +70,19 @@ periods = get_simulation_periods(options_);
params = M_.params;
+if ~isempty(controlled_paths_by_period)
+ for p = 1:periods
+ if isempty(controlled_paths_by_period(p).exogenize_id)
+ continue
+ end
+ endogenousvariables(controlled_paths_by_period(p).exogenize_id, p+M_.maximum_lag) = controlled_paths_by_period(p).values;
+ end
+
+ if options_.debug
+ error('Debugging not available with perfect_foresight_controlled_paths')
+ end
+end
+
if verbose
skipline()
printline(80)
@@ -125,6 +141,10 @@ h2 = clock;
[res, A] = linear_perfect_foresight_problem(Y, jacobian, y0, yT, exogenousvariables, params, steadystate_y, maximum_lag, periods, ny);
+if ~isempty(controlled_paths_by_period)
+ A = controlled_paths_substitute_stacked_jacobian(A, Y, y0, yT, exogenousvariables, steadystate_y, controlled_paths_by_period, M_);
+end
+
% Evaluation of the maximum residual at the initial guess (steady state for the endogenous variables).
err = max(abs(res));
@@ -141,7 +161,8 @@ end
% Try to update the vector of endogenous variables.
try
- Y = Y - A\res;
+ dY = -A\res;
+ Y = Y + dY;
catch
% Normally, because the model is linear, the solution of the perfect foresight model should
% be obtained in one Newton step. This is not the case if the model is singular.
@@ -154,6 +175,18 @@ catch
return
end
+if ~isempty(controlled_paths_by_period)
+ for p = 1:periods
+ endogenize_id = controlled_paths_by_period(p).endogenize_id;
+ exogenize_id = controlled_paths_by_period(p).exogenize_id;
+ if isempty(endogenize_id)
+ continue
+ end
+ Y(exogenize_id+(p-1)*M_.endo_nbr) = controlled_paths_by_period(p).values;
+ exogenousvariables(p+M_.maximum_lag,endogenize_id) = exogenousvariables(p+M_.maximum_lag,endogenize_id) + dY(exogenize_id+(p-1)*M_.endo_nbr)';
+ end
+end
+
YY = reshape([y0; Y; yT], ny, periods+2);
i_rows = (1:ny)';
diff --git a/meson.build b/meson.build
index f52e708e87..351922ff61 100644
--- a/meson.build
+++ b/meson.build
@@ -1385,6 +1385,7 @@ mod_and_m_tests = [
{ 'test' : [ 'deterministic_simulations/Irreversible_investment.mod',
'deterministic_simulations/Irreversible_investment_leadlag.mod' ] },
{ 'test' : [ 'deterministic_simulations/linear_state_space_arma.mod' ] },
+ { 'test' : [ 'deterministic_simulations/linear_state_space_arma_condpaths.mod' ] },
{ 'test' : [ 'deterministic_simulations/purely_forward/ar1.mod' ] },
{ 'test' : [ 'deterministic_simulations/purely_forward/nk.mod' ] },
{ 'test' : [ 'deterministic_simulations/purely_backward/ar1.mod' ] },
diff --git a/tests/deterministic_simulations/linear_state_space_arma_condpaths.mod b/tests/deterministic_simulations/linear_state_space_arma_condpaths.mod
new file mode 100644
index 0000000000..eb408209f6
--- /dev/null
+++ b/tests/deterministic_simulations/linear_state_space_arma_condpaths.mod
@@ -0,0 +1,60 @@
+% Tests perfect simulation of linear model (sim1_linear.m) with controlled paths
+
+var x
+ y
+ z;
+
+varexo u
+ v;
+
+parameters a1 a2 a3 a4
+ b1 b2 b3
+ c1;
+
+a1 = .50;
+a2 = .00;
+a3 = .70;
+a4 = .40;
+b1 = .90;
+b2 = .00;
+b3 = .80;
+c1 = .95;
+
+
+model(linear);
+ y = a1*x(-1) + a2*x(+1) + a3*z + a4*y(-1);
+ z = b1*z(-1) + b2*z(+1) + b3*x + u;
+ x = c1*x(-1) + v +v(-1)+v(+1);
+end;
+
+initval;
+ y=-1;
+ x=-1;
+ z=-1;
+end;
+
+endval;
+ y=0;
+ x=0;
+ z=0;
+end;
+
+perfect_foresight_controlled_paths;
+ exogenize z;
+ periods 50;
+ values 0.5;
+ endogenize u;
+end;
+
+steady;
+
+perfect_foresight_setup(periods=100);
+perfect_foresight_solver;
+
+if ~oo_.deterministic_simulation.status
+ error('Perfect foresight simulation failed')
+end
+
+if ~(oo_.endo_simul(3, 51) == 0.5 && isequal(find(oo_.exo_simul ~= 0), 51))
+ error('Linear perfect foresight simulation with controlled paths failed')
+end
--
GitLab