Select Git revision
-
Houtan Bastani authoredHoutan Bastani authored
DynamicModel.cc 142.81 KiB
/*
* Copyright (C) 2003-2009 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/>.
*/
#include <iostream>
#include <cmath>
#include <cstdlib>
#include <cassert>
#include <cstdio>
#include <cerrno>
#include "DynamicModel.hh"
// For mkdir() and chdir()
#ifdef _WIN32
# include <direct.h>
#else
# include <unistd.h>
# include <sys/stat.h>
# include <sys/types.h>
#endif
DynamicModel::DynamicModel(SymbolTable &symbol_table_arg,
NumericalConstants &num_constants_arg) :
ModelTree(symbol_table_arg, num_constants_arg),
max_lag(0), max_lead(0),
max_endo_lag(0), max_endo_lead(0),
max_exo_lag(0), max_exo_lead(0),
max_exo_det_lag(0), max_exo_det_lead(0),
dynJacobianColsNbr(0),
cutoff(1e-15),
mfs(0),
block_triangular(symbol_table_arg, num_constants_arg)
{
}
VariableNode *
DynamicModel::AddVariable(int symb_id, int lag)
{
return AddVariableInternal(symb_id, lag);
}
void
DynamicModel::compileDerivative(ofstream &code_file, int eq, int symb_id, int lag, map_idx_type &map_idx) const
{
//first_derivatives_type::const_iterator it = first_derivatives.find(make_pair(eq, getDerivID(symb_id, lag)));
first_derivatives_type::const_iterator it = first_derivatives.find(make_pair(eq, getDerivID(symbol_table.getID(eEndogenous, symb_id), lag)));
if (it != first_derivatives.end())
(it->second)->compile(code_file, false, temporary_terms, map_idx, true, false);
else
code_file.write(&FLDZ, sizeof(FLDZ));
}
void
DynamicModel::compileChainRuleDerivative(ofstream &code_file, int eqr, int varr, int lag, map_idx_type &map_idx) const{
map<pair<int, pair<int, int> >, NodeID>::const_iterator it = first_chain_rule_derivatives.find(make_pair(eqr, make_pair(varr, lag)));
if (it != first_chain_rule_derivatives.end())
(it->second)->compile(code_file, false, temporary_terms, map_idx, true, false);
else
code_file.write(&FLDZ, sizeof(FLDZ));
}
void
DynamicModel::BuildIncidenceMatrix()
{
set<pair<int, int> > endogenous, exogenous;
for (int eq = 0; eq < (int) equations.size(); eq++)
{
BinaryOpNode *eq_node = equations[eq];
endogenous.clear();
NodeID Id = eq_node->get_arg1();
Id->collectEndogenous(endogenous);
Id = eq_node->get_arg2();
Id->collectEndogenous(endogenous);
for (set<pair<int, int> >::iterator it_endogenous=endogenous.begin();it_endogenous!=endogenous.end();it_endogenous++)
{
block_triangular.incidencematrix.fill_IM(eq, it_endogenous->first, it_endogenous->second, eEndogenous);
}
exogenous.clear();
Id = eq_node->get_arg1();
Id->collectExogenous(exogenous);
Id = eq_node->get_arg2();
Id->collectExogenous(exogenous);
for (set<pair<int, int> >::iterator it_exogenous=exogenous.begin();it_exogenous!=exogenous.end();it_exogenous++)
{
block_triangular.incidencematrix.fill_IM(eq, it_exogenous->first, it_exogenous->second, eExogenous);
}
}
}
void
DynamicModel::computeTemporaryTermsOrdered(Model_Block *ModelBlock)
{
map<NodeID, pair<int, int> > first_occurence;
map<NodeID, int> reference_count;
int i, j, m, eq, var, eqr, varr, lag;
temporary_terms_type vect;
ostringstream tmp_output;
BinaryOpNode *eq_node;
first_derivatives_type::const_iterator it;
first_chain_rule_derivatives_type::const_iterator it_chr;
ostringstream tmp_s;
temporary_terms.clear();
map_idx.clear();
for (j = 0;j < ModelBlock->Size;j++)
{
// Compute the temporary terms reordered
for (i = 0;i < ModelBlock->Block_List[j].Size;i++)
{
if (ModelBlock->Block_List[j].Equation_Type[i] == E_EVALUATE_S && i<ModelBlock->Block_List[j].Nb_Recursives && ModelBlock->Block_List[j].Equation_Normalized[i])
ModelBlock->Block_List[j].Equation_Normalized[i]->computeTemporaryTerms(reference_count, temporary_terms, first_occurence, j, ModelBlock, i, map_idx);
else
{
eq_node = equations[ModelBlock->Block_List[j].Equation[i]];
eq_node->computeTemporaryTerms(reference_count, temporary_terms, first_occurence, j, ModelBlock, i, map_idx);
}
}
for(i=0; i<(int)ModelBlock->Block_List[j].Chain_Rule_Derivatives->size();i++)
{
pair< pair<int, pair<int, int> >, pair<int, int> > it = ModelBlock->Block_List[j].Chain_Rule_Derivatives->at(i);
lag=it.first.first;
int eqr=it.second.first; int varr=it.second.second;
it_chr=first_chain_rule_derivatives.find(make_pair(eqr, make_pair( varr, lag)));
it_chr->second->computeTemporaryTerms(reference_count, temporary_terms, first_occurence, j, ModelBlock, ModelBlock->Block_List[j].Size-1, map_idx);
}
for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
lag=m-ModelBlock->Block_List[j].Max_Lag;
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size;i++)
{
eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_Index[i];
var=ModelBlock->Block_List[j].IM_lead_lag[m].Var_Index[i];
it=first_derivatives.find(make_pair(eq,getDerivID(symbol_table.getID(eEndogenous, var), lag)));
it->second->computeTemporaryTerms(reference_count, temporary_terms, first_occurence, j, ModelBlock, ModelBlock->Block_List[j].Size-1, map_idx);
}
}
/*for(i=0; i<(int)ModelBlock->Block_List[j].Chain_Rule_Derivatives->size();i++)
{
pair< pair<int, pair<int, int> >, pair<int, int> > it = ModelBlock->Block_List[j].Chain_Rule_Derivatives->at(i);
lag=it.first.first;
eqr=it.second.first;
varr=it.second.second;
it_chr=first_chain_rule_derivatives.find(make_pair(eqr, make_pair( varr, lag)));
it_chr->second->computeTemporaryTerms(reference_count, temporary_terms, first_occurence, j, ModelBlock, ModelBlock->Block_List[j].Size-1, map_idx);
}*/
/*for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
lag=m-ModelBlock->Block_List[j].Max_Lag;
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size_exo;i++)
{
eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_X_Index[i];
var=ModelBlock->Block_List[j].IM_lead_lag[m].Exogenous_Index[i];
it=first_derivatives.find(make_pair(eq,getDerivID(symbol_table.getID(eExogenous, var), lag)));
it->second->computeTemporaryTerms(reference_count, temporary_terms, first_occurence, j, ModelBlock, ModelBlock->Block_List[j].Size-1, map_idx);
}
}*/
//jacobian_max_exo_col=(variable_table.max_exo_lag+variable_table.max_exo_lead+1)*symbol_table.exo_nbr;
for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
lag=m-ModelBlock->Block_List[j].Max_Lag;
if (block_triangular.incidencematrix.Model_Max_Lag_Endo - ModelBlock->Block_List[j].Max_Lag +m >=0)
{
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size_other_endo;i++)
{
eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_Index_other_endo[i];
var=ModelBlock->Block_List[j].IM_lead_lag[m].Var_Index_other_endo[i];
it=first_derivatives.find(make_pair(eq,getDerivID(symbol_table.getID(eEndogenous, var), lag)));
it->second->computeTemporaryTerms(reference_count, temporary_terms, first_occurence, j, ModelBlock, ModelBlock->Block_List[j].Size-1, map_idx);
}
}
}
}
for (j = 0;j < ModelBlock->Size;j++)
{
// Collecte the temporary terms reordered
for (i = 0;i < ModelBlock->Block_List[j].Size;i++)
{
if (ModelBlock->Block_List[j].Equation_Type[i] == E_EVALUATE_S && i<ModelBlock->Block_List[j].Nb_Recursives && ModelBlock->Block_List[j].Equation_Normalized[i])
ModelBlock->Block_List[j].Equation_Normalized[i]->collectTemporary_terms(temporary_terms, ModelBlock, j);
else
{
eq_node = equations[ModelBlock->Block_List[j].Equation[i]];
eq_node->collectTemporary_terms(temporary_terms, ModelBlock, j);
}
/*eq_node = equations[ModelBlock->Block_List[j].Equation[i]];
eq_node->collectTemporary_terms(temporary_terms, ModelBlock, j);
if (ModelBlock->Block_List[j].Equation_Type[i] == E_EVALUATE_S)
if(ModelBlock->Block_List[j].Equation_Normalized[i])
ModelBlock->Block_List[j].Equation_Normalized[i]->collectTemporary_terms(temporary_terms, ModelBlock, j); for(temporary_terms_type::const_iterator it = ModelBlock->Block_List[j].Temporary_Terms_in_Equation[i]->begin(); it!= ModelBlock->Block_List[j].Temporary_Terms_in_Equation[i]->end(); it++)
(*it)->collectTemporary_terms(temporary_terms, ModelBlock, j);*/
}
for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
lag=m-ModelBlock->Block_List[j].Max_Lag;
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size;i++)
{
eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_Index[i];
var=ModelBlock->Block_List[j].IM_lead_lag[m].Var_Index[i];
it=first_derivatives.find(make_pair(eq,getDerivID(symbol_table.getID(eEndogenous, var), lag)));
//it=first_derivatives.find(make_pair(eq,variable_table.getID(var, lag)));
//if(it!=first_derivatives.end())
it->second->collectTemporary_terms(temporary_terms, ModelBlock, j);
}
}
for(i=0; i<(int)ModelBlock->Block_List[j].Chain_Rule_Derivatives->size();i++)
{
pair< pair<int, pair<int, int> >, pair<int, int> > it = ModelBlock->Block_List[j].Chain_Rule_Derivatives->at(i);
lag=it.first.first;
eqr=it.second.first;
varr=it.second.second;
it_chr=first_chain_rule_derivatives.find(make_pair(eqr, make_pair( varr, lag)));
it_chr->second->collectTemporary_terms(temporary_terms, ModelBlock, j);
}
/*for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
lag=m-ModelBlock->Block_List[j].Max_Lag;
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size_exo;i++)
{
eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_X_Index[i];
var=ModelBlock->Block_List[j].IM_lead_lag[m].Exogenous_Index[i];
it=first_derivatives.find(make_pair(eq,getDerivID(symbol_table.getID(eExogenous, var), lag)));
//it=first_derivatives.find(make_pair(eq,variable_table.getID(var, lag)));
it->second->collectTemporary_terms(temporary_terms, ModelBlock, j);
}
}*/
//jacobian_max_exo_col=(variable_table.max_exo_lag+variable_table.max_exo_lead+1)*symbol_table.exo_nbr;
for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
lag=m-ModelBlock->Block_List[j].Max_Lag;
if (block_triangular.incidencematrix.Model_Max_Lag_Endo - ModelBlock->Block_List[j].Max_Lag +m >=0)
{
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size_other_endo;i++)
{
eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_Index_other_endo[i];
var=ModelBlock->Block_List[j].IM_lead_lag[m].Var_Index_other_endo[i];
it=first_derivatives.find(make_pair(eq,getDerivID(symbol_table.getID(eEndogenous, var), lag)));
//it=first_derivatives.find(make_pair(eq,variable_table.getID(var, lag)));
//if(it!=first_derivatives.end())
it->second->collectTemporary_terms(temporary_terms, ModelBlock, j);
}
}
}
}
// Add a mapping form node ID to temporary terms order
j=0;
for (temporary_terms_type::const_iterator it = temporary_terms.begin();
it != temporary_terms.end(); it++)
map_idx[(*it)->idx]=j++;
}
void
DynamicModel::writeModelEquationsOrdered_M( Model_Block *ModelBlock, const string &dynamic_basename) const
{
int i,j,k,m;
string tmp_s, sps;
ostringstream tmp_output, tmp1_output, global_output;
NodeID lhs=NULL, rhs=NULL;
BinaryOpNode *eq_node; ostringstream Uf[symbol_table.endo_nbr()];
map<NodeID, int> reference_count;
//int prev_Simulation_Type=-1, count_derivates=0;
int jacobian_max_endo_col;
ofstream output;
//temporary_terms_type::const_iterator it_temp=temporary_terms.begin();
int nze, nze_exo, nze_other_endo;
//map<int, NodeID> recursive_variables;
vector<int> feedback_variables;
//----------------------------------------------------------------------
//For each block
for (j = 0;j < ModelBlock->Size;j++)
{
//recursive_variables.clear();
feedback_variables.clear();
//For a block composed of a single equation determines wether we have to evaluate or to solve the equation
nze = nze_exo = nze_other_endo = 0;
for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
nze+=ModelBlock->Block_List[j].IM_lead_lag[m].size;
/*for (m=0;m<=ModelBlock->Block_List[j].Max_Lead_Exo+ModelBlock->Block_List[j].Max_Lag_Exo;m++)
nze_exo+=ModelBlock->Block_List[j].IM_lead_lag[m].size_exo;*/
for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
k=m-ModelBlock->Block_List[j].Max_Lag;
if (block_triangular.incidencematrix.Model_Max_Lag_Endo - ModelBlock->Block_List[j].Max_Lag +m >=0)
nze_other_endo+=ModelBlock->Block_List[j].IM_lead_lag[m].size_other_endo;
}
tmp1_output.str("");
tmp1_output << dynamic_basename << "_" << j+1 << ".m";
output.open(tmp1_output.str().c_str(), ios::out | ios::binary);
output << "%\n";
output << "% " << tmp1_output.str() << " : Computes dynamic model for Dynare\n";
output << "%\n";
output << "% Warning : this file is generated automatically by Dynare\n";
output << "% from model file (.mod)\n\n";
output << "%/\n";
if (ModelBlock->Block_List[j].Simulation_Type==EVALUATE_BACKWARD
||ModelBlock->Block_List[j].Simulation_Type==EVALUATE_FORWARD
/*||ModelBlock->Block_List[j].Simulation_Type==EVALUATE_BACKWARD_R
||ModelBlock->Block_List[j].Simulation_Type==EVALUATE_FORWARD_R*/)
{
output << "function [y, g1, g2, g3, varargout] = " << dynamic_basename << "_" << j+1 << "(y, x, params, jacobian_eval, y_kmin, periods)\n";
}
else if (ModelBlock->Block_List[j].Simulation_Type==SOLVE_FORWARD_COMPLETE
|| ModelBlock->Block_List[j].Simulation_Type==SOLVE_BACKWARD_COMPLETE)
output << "function [residual, y, g1, g2, g3, varargout] = " << dynamic_basename << "_" << j+1 << "(y, x, params, it_, jacobian_eval)\n";
else if (ModelBlock->Block_List[j].Simulation_Type==SOLVE_BACKWARD_SIMPLE
|| ModelBlock->Block_List[j].Simulation_Type==SOLVE_FORWARD_SIMPLE)
output << "function [residual, y, g1, g2, g3, varargout] = " << dynamic_basename << "_" << j+1 << "(y, x, params, it_, jacobian_eval)\n";
else
output << "function [residual, y, g1, g2, g3, b, varargout] = " << dynamic_basename << "_" << j+1 << "(y, x, params, periods, jacobian_eval, y_kmin, y_size)\n";
output << " % ////////////////////////////////////////////////////////////////////////" << endl
<< " % //" << string(" Block ").substr(int(log10(j + 1))) << j + 1 << " " << BlockTriangular::BlockType0(ModelBlock->Block_List[j].Type)
<< " //" << endl
<< " % // Simulation type "
<< BlockTriangular::BlockSim(ModelBlock->Block_List[j].Simulation_Type) << " //" << endl
<< " % ////////////////////////////////////////////////////////////////////////" << endl;
output << " global options_;" << endl;
//The Temporary terms
//output << " relax = 1;\n";
if (ModelBlock->Block_List[j].Simulation_Type==EVALUATE_BACKWARD
||ModelBlock->Block_List[j].Simulation_Type==EVALUATE_FORWARD)
{
output << " if(jacobian_eval)\n";
output << " g1 = spalloc(" << ModelBlock->Block_List[j].Size-ModelBlock->Block_List[j].Nb_Recursives
<< ", " << (ModelBlock->Block_List[j].Size-ModelBlock->Block_List[j].Nb_Recursives)*(1+ModelBlock->Block_List[j].Max_Lag_Endo+ModelBlock->Block_List[j].Max_Lead_Endo)
<< ", " << nze << ");\n";
output << " g1_x=spalloc(" << ModelBlock->Block_List[j].Size << ", " << (ModelBlock->Block_List[j].nb_exo + ModelBlock->Block_List[j].nb_exo_det)*(1+ModelBlock->Block_List[j].Max_Lag_Exo+ModelBlock->Block_List[j].Max_Lead_Exo) << ", " << nze_exo << ");\n";
output << " g1_o=spalloc(" << ModelBlock->Block_List[j].Size << ", " << ModelBlock->Block_List[j].nb_other_endo*(1+ModelBlock->Block_List[j].Max_Lag_Other_Endo+ModelBlock->Block_List[j].Max_Lead_Other_Endo) << ", " << nze_other_endo << ");\n";
output << " end;\n"; }
else
{
output << " if(jacobian_eval)\n";
output << " g1 = spalloc(" << ModelBlock->Block_List[j].Size << ", " << ModelBlock->Block_List[j].Size*(1+ModelBlock->Block_List[j].Max_Lag_Endo+ModelBlock->Block_List[j].Max_Lead_Endo) << ", " << nze << ");\n";
output << " g1_x=spalloc(" << ModelBlock->Block_List[j].Size << ", " << (ModelBlock->Block_List[j].nb_exo + ModelBlock->Block_List[j].nb_exo_det)*(1+ModelBlock->Block_List[j].Max_Lag_Exo+ModelBlock->Block_List[j].Max_Lead_Exo) << ", " << nze_exo << ");\n";
output << " g1_o=spalloc(" << ModelBlock->Block_List[j].Size << ", " << ModelBlock->Block_List[j].nb_other_endo*(1+ModelBlock->Block_List[j].Max_Lag_Other_Endo+ModelBlock->Block_List[j].Max_Lead_Other_Endo) << ", " << nze_other_endo << ");\n";
output << " else\n";
if (ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_COMPLETE || ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_SIMPLE)
{
output << " g1 = spalloc(" << (ModelBlock->Block_List[j].Size-ModelBlock->Block_List[j].Nb_Recursives) << "*options_.periods, "
<< (ModelBlock->Block_List[j].Size-ModelBlock->Block_List[j].Nb_Recursives) << "*(options_.periods+" << ModelBlock->Block_List[j].Max_Lag+ModelBlock->Block_List[j].Max_Lead+1 << ")"
<< ", " << nze << "*options_.periods);\n";
/*output << " g1_tmp_r = spalloc(" << (ModelBlock->Block_List[j].Nb_Recursives)
<< ", " << (ModelBlock->Block_List[j].Size)*(ModelBlock->Block_List[j].Max_Lag+ModelBlock->Block_List[j].Max_Lead+1)
<< ", " << nze << ");\n";
output << " g1_tmp_b = spalloc(" << (ModelBlock->Block_List[j].Size-ModelBlock->Block_List[j].Nb_Recursives)
<< ", " << (ModelBlock->Block_List[j].Size)*(ModelBlock->Block_List[j].Max_Lag+ModelBlock->Block_List[j].Max_Lead+1)
<< ", " << nze << ");\n";*/
}
else
{
output << " g1 = spalloc(" << ModelBlock->Block_List[j].Size-ModelBlock->Block_List[j].Nb_Recursives
<< ", " << ModelBlock->Block_List[j].Size-ModelBlock->Block_List[j].Nb_Recursives << ", " << nze << ");\n";
output << " g1_tmp_r = spalloc(" << ModelBlock->Block_List[j].Nb_Recursives
<< ", " << ModelBlock->Block_List[j].Size << ", " << nze << ");\n";
output << " g1_tmp_b = spalloc(" << ModelBlock->Block_List[j].Size-ModelBlock->Block_List[j].Nb_Recursives
<< ", " << ModelBlock->Block_List[j].Size << ", " << nze << ");\n";
}
output << " end;\n";
}
output << " g2=0;g3=0;\n";
if (ModelBlock->Block_List[j].Temporary_InUse->size())
{
tmp_output.str("");
for (temporary_terms_inuse_type::const_iterator it = ModelBlock->Block_List[j].Temporary_InUse->begin();
it != ModelBlock->Block_List[j].Temporary_InUse->end(); it++)
tmp_output << " T" << *it;
output << " global" << tmp_output.str() << ";\n";
}
if (ModelBlock->Block_List[j].Simulation_Type!=EVALUATE_BACKWARD && ModelBlock->Block_List[j].Simulation_Type!=EVALUATE_FORWARD)
output << " residual=zeros(" << ModelBlock->Block_List[j].Size-ModelBlock->Block_List[j].Nb_Recursives << ",1);\n";
if (ModelBlock->Block_List[j].Simulation_Type==EVALUATE_BACKWARD)
output << " for it_ = (y_kmin+periods):y_kmin+1\n";
if (ModelBlock->Block_List[j].Simulation_Type==EVALUATE_FORWARD)
output << " for it_ = y_kmin+1:(y_kmin+periods)\n";
if (ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_COMPLETE || ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_SIMPLE)
{
output << " b = zeros(periods*y_size,1);\n";
output << " for it_ = y_kmin+1:(periods+y_kmin)\n";
output << " Per_y_=it_*y_size;\n";
output << " Per_J_=(it_-y_kmin-1)*y_size;\n";
output << " Per_K_=(it_-1)*y_size;\n";
sps=" ";
}
else
if (ModelBlock->Block_List[j].Simulation_Type==EVALUATE_BACKWARD || ModelBlock->Block_List[j].Simulation_Type==EVALUATE_FORWARD )
sps = " ";
else
sps="";
// The equations
for (i = 0;i < ModelBlock->Block_List[j].Size;i++)
{
temporary_terms_type tt2;
tt2.clear();
if (ModelBlock->Block_List[j].Temporary_Terms_in_Equation[i]->size())
output << " " << sps << "% //Temporary variables" << endl;
for (temporary_terms_type::const_iterator it = ModelBlock->Block_List[j].Temporary_Terms_in_Equation[i]->begin(); it != ModelBlock->Block_List[j].Temporary_Terms_in_Equation[i]->end(); it++)
{
output << " " << sps;
(*it)->writeOutput(output, oMatlabDynamicModelSparse, temporary_terms);
output << " = ";
(*it)->writeOutput(output, oMatlabDynamicModelSparse, tt2);
// Insert current node into tt2
tt2.insert(*it);
output << ";" << endl;
}
string sModel = symbol_table.getName(symbol_table.getID(eEndogenous, ModelBlock->Block_List[j].Variable[i])) ;
eq_node = equations[ModelBlock->Block_List[j].Equation[i]];
lhs = eq_node->get_arg1();
rhs = eq_node->get_arg2();
tmp_output.str("");
/*if((ModelBlock->Block_List[j].Simulation_Type!=EVALUATE_BACKWARD or ModelBlock->Block_List[j].Simulation_Type!=EVALUATE_FORWARD) and (i<ModelBlock->Block_List[j].Nb_Recursives))
lhs->writeOutput(tmp_output, oMatlabDynamicModelSparse, temporary_terms);
else*/
lhs->writeOutput(tmp_output, oMatlabDynamicModelSparse, temporary_terms);
switch (ModelBlock->Block_List[j].Simulation_Type)
{
case EVALUATE_BACKWARD:
case EVALUATE_FORWARD:
evaluation: if (ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_COMPLETE || ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_SIMPLE)
output << " % equation " << ModelBlock->Block_List[j].Equation[i]+1 << " variable : " << sModel
<< " (" << ModelBlock->Block_List[j].Variable[i]+1 << ") " << block_triangular.c_Equation_Type(ModelBlock->Block_List[j].Equation_Type[i]) << endl;
output << " ";
if (ModelBlock->Block_List[j].Equation_Type[i] == E_EVALUATE)
{
output << tmp_output.str();
output << " = ";
/*if(!(ModelBlock->Block_List[j].Simulation_Type==EVALUATE_BACKWARD or ModelBlock->Block_List[j].Simulation_Type==EVALUATE_FORWARD))
{
lhs->writeOutput(output, oMatlabDynamicModelSparse, temporary_terms);
output << "-relax*(";
lhs->writeOutput(output, oMatlabDynamicModelSparse, temporary_terms);
output << "-(";
rhs->writeOutput(output, oMatlabDynamicModelSparse, temporary_terms);
output << "))";
}
else*/
rhs->writeOutput(output, oMatlabDynamicModelSparse, temporary_terms);
}
else if (ModelBlock->Block_List[j].Equation_Type[i] == E_EVALUATE_S)
{
output << "%" << tmp_output.str();
output << " = ";
if (ModelBlock->Block_List[j].Equation_Normalized[i])
{
rhs->writeOutput(output, oMatlabDynamicModelSparse, temporary_terms);
output << "\n ";
tmp_output.str("");
eq_node = (BinaryOpNode *)ModelBlock->Block_List[j].Equation_Normalized[i];
lhs = eq_node->get_arg1();
rhs = eq_node->get_arg2();
lhs->writeOutput(output, oMatlabDynamicModelSparse, temporary_terms);
output << " = ";
/*if(!(ModelBlock->Block_List[j].Simulation_Type==EVALUATE_BACKWARD or ModelBlock->Block_List[j].Simulation_Type==EVALUATE_FORWARD))
{
lhs->writeOutput(output, oMatlabDynamicModelSparse, temporary_terms);
output << "-relax*(";
lhs->writeOutput(output, oMatlabDynamicModelSparse, temporary_terms);
output << "-(";
rhs->writeOutput(output, oMatlabDynamicModelSparse, temporary_terms);
output << "))";
}
else*/
rhs->writeOutput(output, oMatlabDynamicModelSparse, temporary_terms);
}
} else
{
cerr << "Type missmatch for equation " << ModelBlock->Block_List[j].Equation[i]+1 << "\n";
exit(EXIT_FAILURE);
}
output << ";\n";
break;
case SOLVE_BACKWARD_SIMPLE:
case SOLVE_FORWARD_SIMPLE:
case SOLVE_BACKWARD_COMPLETE:
case SOLVE_FORWARD_COMPLETE:
if (i<ModelBlock->Block_List[j].Nb_Recursives)
{
/*if (ModelBlock->Block_List[j].Equation_Type[i] == E_EVALUATE_S)
recursive_variables[getDerivID(symbol_table.getID(eEndogenous, ModelBlock->Block_List[j].Variable[i]), 0)] = ModelBlock->Block_List[j].Equation_Normalized[i];
else
recursive_variables[getDerivID(symbol_table.getID(eEndogenous, ModelBlock->Block_List[j].Variable[i]), 0)] = equations[ModelBlock->Block_List[j].Equation[i]];*/
goto evaluation;
}
feedback_variables.push_back(ModelBlock->Block_List[j].Variable[i]);
output << " % equation " << ModelBlock->Block_List[j].Equation[i]+1 << " variable : " << sModel
<< " (" << ModelBlock->Block_List[j].Variable[i]+1 << ") " << block_triangular.c_Equation_Type(ModelBlock->Block_List[j].Equation_Type[i]) << endl;
output << " " << "residual(" << i+1-ModelBlock->Block_List[j].Nb_Recursives << ") = (";
goto end;
case SOLVE_TWO_BOUNDARIES_COMPLETE:
case SOLVE_TWO_BOUNDARIES_SIMPLE:
if (i<ModelBlock->Block_List[j].Nb_Recursives)
{
/*if (ModelBlock->Block_List[j].Equation_Type[i] == E_EVALUATE_S)
recursive_variables[getDerivID(symbol_table.getID(eEndogenous, ModelBlock->Block_List[j].Variable[i]), 0)] = ModelBlock->Block_List[j].Equation_Normalized[i];
else
recursive_variables[getDerivID(symbol_table.getID(eEndogenous, ModelBlock->Block_List[j].Variable[i]), 0)] = equations[ModelBlock->Block_List[j].Equation[i]];*/
goto evaluation;
}
feedback_variables.push_back(ModelBlock->Block_List[j].Variable[i]);
output << " % equation " << ModelBlock->Block_List[j].Equation[i]+1 << " variable : " << sModel
<< " (" << ModelBlock->Block_List[j].Variable[i]+1 << ") " << block_triangular.c_Equation_Type(ModelBlock->Block_List[j].Equation_Type[i]) << endl;
Uf[ModelBlock->Block_List[j].Equation[i]] << " b(" << i+1-ModelBlock->Block_List[j].Nb_Recursives << "+Per_J_) = -residual(" << i+1-ModelBlock->Block_List[j].Nb_Recursives << ", it_)";
output << " residual(" << i+1-ModelBlock->Block_List[j].Nb_Recursives << ", it_) = (";
goto end;
default:
end:
output << tmp_output.str();
output << ") - (";
rhs->writeOutput(output, oMatlabDynamicModelSparse, temporary_terms);
output << ");\n";
#ifdef CONDITION
if (ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_COMPLETE || ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_SIMPLE)
output << " condition(" << i+1 << ")=0;\n";
#endif
}
}
// The Jacobian if we have to solve the block
if (ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_SIMPLE
|| ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_COMPLETE)
output << " " << sps << "% Jacobian " << endl;
else
if (ModelBlock->Block_List[j].Simulation_Type==SOLVE_BACKWARD_SIMPLE || ModelBlock->Block_List[j].Simulation_Type==SOLVE_FORWARD_SIMPLE ||
ModelBlock->Block_List[j].Simulation_Type==SOLVE_BACKWARD_COMPLETE || ModelBlock->Block_List[j].Simulation_Type==SOLVE_FORWARD_COMPLETE)
output << " % Jacobian " << endl << " if jacobian_eval" << endl;
else
output << " % Jacobian " << endl << " if jacobian_eval" << endl;
switch (ModelBlock->Block_List[j].Simulation_Type)
{
case EVALUATE_BACKWARD:
case EVALUATE_FORWARD:
for (m=0;m<ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag+1;m++)
{
k=m-ModelBlock->Block_List[j].Max_Lag;
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size;i++) {
int eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_Index[i];
int var=ModelBlock->Block_List[j].IM_lead_lag[m].Var_Index[i];
int eqr=ModelBlock->Block_List[j].IM_lead_lag[m].Equ[i];
int varr=ModelBlock->Block_List[j].IM_lead_lag[m].Var[i];
output << " g1(" << eqr+1 << ", " << /*varr+1+(m+variable_table.max_lag-ModelBlock->Block_List[j].Max_Lag)*symbol_table.endo_nbr*/
varr+1+m*ModelBlock->Block_List[j].Size << ") = ";
writeDerivative(output, eq, symbol_table.getID(eEndogenous, var), k, oMatlabDynamicModelSparse, temporary_terms);
output << "; % variable=" << symbol_table.getName(symbol_table.getID(eEndogenous, var))
<< "(" << k//variable_table.getLag(variable_table.getSymbolID(ModelBlock->Block_List[j].Variable[0]))
<< ") " << var+1
<< ", equation=" << eq+1 << endl;
}
}
//jacobian_max_endo_col=(variable_table.max_endo_lag+variable_table.max_endo_lead+1)*symbol_table.endo_nbr;
/*for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
k=m-ModelBlock->Block_List[j].Max_Lag;
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size_exo;i++)
{
int eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_X_Index[i];
int var=ModelBlock->Block_List[j].IM_lead_lag[m].Exogenous_Index[i];
int eqr=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_X[i];
int varr=ModelBlock->Block_List[j].IM_lead_lag[m].Exogenous[i];
output << " g1_x(" << eqr+1 << ", "
<< varr+1+(m+max_exo_lag-ModelBlock->Block_List[j].Max_Lag)*symbol_table.exo_nbr() << ") = ";
writeDerivative(output, eq, symbol_table.getID(eExogenous, var), k, oMatlabDynamicModelSparse, temporary_terms);
output << "; % variable=" << symbol_table.getName(var)
<< "(" << k << ") " << var+1
<< ", equation=" << eq+1 << endl;
}
}*/
for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
k=m-ModelBlock->Block_List[j].Max_Lag;
if (block_triangular.incidencematrix.Model_Max_Lag_Endo - ModelBlock->Block_List[j].Max_Lag +m >=0)
{
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size_other_endo;i++)
{
int eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_Index_other_endo[i];
int var=ModelBlock->Block_List[j].IM_lead_lag[m].Var_Index_other_endo[i];
int eqr=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_other_endo[i];
int varr=ModelBlock->Block_List[j].IM_lead_lag[m].Var_other_endo[i];
output << " g1_o(" << eqr+1 << ", "
<< varr+1+(m+max_endo_lag-ModelBlock->Block_List[j].Max_Lag)*symbol_table.endo_nbr() << ") = ";
writeDerivative(output, eq, symbol_table.getID(eEndogenous, var), k, oMatlabDynamicModelSparse, temporary_terms);
output << "; % variable=" << symbol_table.getName(symbol_table.getID(eEndogenous, var))
<< "(" << k << ") " << var+1
<< ", equation=" << eq+1 << endl;
}
}
}
output << " varargout{1}=g1_x;\n";
output << " varargout{2}=g1_o;\n";
output << " end;" << endl;
//output << " ya = y;\n";
output << " end;" << endl;
break;
case SOLVE_BACKWARD_SIMPLE:
case SOLVE_FORWARD_SIMPLE:
case SOLVE_BACKWARD_COMPLETE:
case SOLVE_FORWARD_COMPLETE:
for (m=0;m<ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag+1;m++)
{
k=m-ModelBlock->Block_List[j].Max_Lag;
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size;i++)
{
int eqr=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_Index[i];
int varr=ModelBlock->Block_List[j].IM_lead_lag[m].Var_Index[i];
int eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ[i]; int var=ModelBlock->Block_List[j].IM_lead_lag[m].Var[i];
output << " g1(" << eq+1 << ", "
<< var+1 + m*(ModelBlock->Block_List[j].Size) << ") = ";
writeDerivative(output, eqr, symbol_table.getID(eEndogenous, varr), k, oMatlabDynamicModelSparse, temporary_terms);
output << "; % variable=" << symbol_table.getName(symbol_table.getID(eEndogenous, varr))
<< "(" << k << ") " << varr+1
<< ", equation=" << eqr+1 << endl;
}
}
/*for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
k=m-ModelBlock->Block_List[j].Max_Lag;
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size_exo;i++)
{
int eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_X_Index[i];
int var=ModelBlock->Block_List[j].IM_lead_lag[m].Exogenous_Index[i];
int eqr=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_X[i];
int varr=ModelBlock->Block_List[j].IM_lead_lag[m].Exogenous[i];
output << " g1_x(" << eqr+1 << ", " << varr+1+(m+max_exo_lag-ModelBlock->Block_List[j].Max_Lag)*ModelBlock->Block_List[j].nb_exo << ") = ";
writeDerivative(output, eq, symbol_table.getID(eExogenous, var), k, oMatlabDynamicModelSparse, temporary_terms);
output << "; % variable=" << symbol_table.getName(var)
<< "(" << k << ") " << var+1
<< ", equation=" << eq+1 << endl;
}
}*/
for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
k=m-ModelBlock->Block_List[j].Max_Lag;
if (block_triangular.incidencematrix.Model_Max_Lag_Endo - ModelBlock->Block_List[j].Max_Lag +m >=0)
{
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size_other_endo;i++)
{
int eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_Index_other_endo[i];
int var=ModelBlock->Block_List[j].IM_lead_lag[m].Var_Index_other_endo[i];
int eqr=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_other_endo[i];
int varr=ModelBlock->Block_List[j].IM_lead_lag[m].Var_other_endo[i];
output << " g1_o(" << eqr+1/*-ModelBlock->Block_List[j].Nb_Recursives*/ << ", "
<< varr+1+(m+max_endo_lag-ModelBlock->Block_List[j].Max_Lag)*symbol_table.endo_nbr() << ") = ";
writeDerivative(output, eq, symbol_table.getID(eEndogenous, var), k, oMatlabDynamicModelSparse, temporary_terms);
output << "; % variable=" << symbol_table.getName(symbol_table.getID(eEndogenous, var))
<< "(" << k << ") " << var+1
<< ", equation=" << eq+1 << endl;
}
}
}
output << " varargout{1}=g1_x;\n";
output << " varargout{2}=g1_o;\n";
output << " else" << endl;
for(i=0; i<(int)ModelBlock->Block_List[j].Chain_Rule_Derivatives->size();i++)
{
pair< pair<int, pair<int, int> >, pair<int, int> > it = ModelBlock->Block_List[j].Chain_Rule_Derivatives->at(i);
k=it.first.first;
int eq=it.first.second.first;
int var=it.first.second.second;
int eqr=it.second.first;
int varr=it.second.second;
output << " g1(" << eq+1-ModelBlock->Block_List[j].Nb_Recursives << ", "
<< var+1-ModelBlock->Block_List[j].Nb_Recursives << ") = ";
writeChainRuleDerivative(output, eqr, varr, k, oMatlabDynamicModelSparse, temporary_terms);
output << "; %2 variable=" << symbol_table.getName(symbol_table.getID(eEndogenous, varr))
<< "(" << k << ") " << varr+1 << ", equation=" << eqr+1 << endl;
}
output << " end;\n";
break;
case SOLVE_TWO_BOUNDARIES_SIMPLE:
case SOLVE_TWO_BOUNDARIES_COMPLETE:
output << " if ~jacobian_eval" << endl;
for(i=0; i<(int)ModelBlock->Block_List[j].Chain_Rule_Derivatives->size();i++)
{ pair< pair<int, pair<int, int> >, pair<int, int> > it = ModelBlock->Block_List[j].Chain_Rule_Derivatives->at(i);
k=it.first.first;
int eq=it.first.second.first;
int var=it.first.second.second;
int eqr=it.second.first;
int varr=it.second.second;
ostringstream tmp_output;
if(eq>=ModelBlock->Block_List[j].Nb_Recursives and var>=ModelBlock->Block_List[j].Nb_Recursives)
{
if (k==0)
Uf[ModelBlock->Block_List[j].Equation[eq]] << "+g1(" << eq+1-ModelBlock->Block_List[j].Nb_Recursives
<< "+Per_J_, " << var+1-ModelBlock->Block_List[j].Nb_Recursives
<< "+Per_K_)*y(it_, " << varr+1 << ")";
else if (k==1)
Uf[ModelBlock->Block_List[j].Equation[eq]] << "+g1(" << eq+1-ModelBlock->Block_List[j].Nb_Recursives
<< "+Per_J_, " << var+1-ModelBlock->Block_List[j].Nb_Recursives
<< "+Per_y_)*y(it_+1, " << varr+1 << ")";
else if (k>0)
Uf[ModelBlock->Block_List[j].Equation[eq]] << "+g1(" << eq+1-ModelBlock->Block_List[j].Nb_Recursives
<< "+Per_J_, " << var+1-ModelBlock->Block_List[j].Nb_Recursives
<< "+y_size*(it_+" << k-1 << "))*y(it_+" << k << ", " << varr+1 << ")";
else if (k<0)
Uf[ModelBlock->Block_List[j].Equation[eq]] << "+g1(" << eq+1-ModelBlock->Block_List[j].Nb_Recursives
<< "+Per_J_, " << var+1-ModelBlock->Block_List[j].Nb_Recursives
<< "+y_size*(it_" << k-1 << "))*y(it_" << k << ", " << varr+1 << ")";
if (k==0)
tmp_output << " g1(" << eq+1-ModelBlock->Block_List[j].Nb_Recursives << "+Per_J_, "
<< var+1-ModelBlock->Block_List[j].Nb_Recursives << "+Per_K_) = ";
else if (k==1)
tmp_output << " g1(" << eq+1-ModelBlock->Block_List[j].Nb_Recursives << "+Per_J_, "
<< var+1-ModelBlock->Block_List[j].Nb_Recursives << "+Per_y_) = ";
else if (k>0)
tmp_output << " g1(" << eq+1-ModelBlock->Block_List[j].Nb_Recursives << "+Per_J_, "
<< var+1-ModelBlock->Block_List[j].Nb_Recursives << "+y_size*(it_+" << k-1 << ")) = ";
else if (k<0)
tmp_output << " g1(" << eq+1-ModelBlock->Block_List[j].Nb_Recursives << "+Per_J_, "
<< var+1-ModelBlock->Block_List[j].Nb_Recursives << "+y_size*(it_" << k-1 << ")) = ";
output << " " << tmp_output.str();
writeChainRuleDerivative(output, eqr, varr, k, oMatlabDynamicModelSparse, temporary_terms);
output << ";";
output << " %2 variable=" << symbol_table.getName(symbol_table.getID(eEndogenous, varr))
<< "(" << k << ") " << varr+1
<< ", equation=" << eqr+1 << " (" << eq+1 << ")" << endl;
}
#ifdef CONDITION
output << " if (fabs(condition[" << eqr << "])<fabs(u[" << u << "+Per_u_]))\n";
output << " condition(" << eqr << ")=u(" << u << "+Per_u_);\n";
#endif
//}
}
for (i = 0;i < ModelBlock->Block_List[j].Size;i++)
{
if (i>=ModelBlock->Block_List[j].Nb_Recursives)
output << " " << Uf[ModelBlock->Block_List[j].Equation[i]].str() << ";\n";
#ifdef CONDITION
output << " if (fabs(condition(" << i+1 << "))<fabs(u(" << i << "+Per_u_)))\n";
output << " condition(" << i+1 << ")=u(" << i+1 << "+Per_u_);\n";
#endif
}
#ifdef CONDITION
for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
k=m-ModelBlock->Block_List[j].Max_Lag;
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size;i++)
{
int eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_Index[i];
int var=ModelBlock->Block_List[j].IM_lead_lag[m].Var_Index[i];
int u=ModelBlock->Block_List[j].IM_lead_lag[m].u[i]; int eqr=ModelBlock->Block_List[j].IM_lead_lag[m].Equ[i];
output << " u(" << u+1 << "+Per_u_) = u(" << u+1 << "+Per_u_) / condition(" << eqr+1 << ");\n";
}
}
for (i = 0;i < ModelBlock->Block_List[j].Size;i++)
output << " u(" << i+1 << "+Per_u_) = u(" << i+1 << "+Per_u_) / condition(" << i+1 << ");\n";
#endif
output << " else" << endl;
for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
k=m-ModelBlock->Block_List[j].Max_Lag;
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size;i++)
{
int eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_Index[i];
int var=ModelBlock->Block_List[j].IM_lead_lag[m].Var_Index[i];
int eqr=ModelBlock->Block_List[j].IM_lead_lag[m].Equ[i];
int varr=ModelBlock->Block_List[j].IM_lead_lag[m].Var[i];
output << " g1(" << eqr+1 << ", " << varr+1+(m-ModelBlock->Block_List[j].Max_Lag+ModelBlock->Block_List[j].Max_Lag_Endo)*ModelBlock->Block_List[j].Size << ") = ";
writeDerivative(output, eq, symbol_table.getID(eEndogenous, var), k, oMatlabDynamicModelSparse, temporary_terms);
output << "; % variable=" << symbol_table.getName(symbol_table.getID(eEndogenous, var))
<< "(" << k << ") " << var+1
<< ", equation=" << eq+1 << endl;
}
}
jacobian_max_endo_col=(ModelBlock->Block_List[j].Max_Lead_Endo+ModelBlock->Block_List[j].Max_Lag_Endo+1)*ModelBlock->Block_List[j].Size;
/*for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
k=m-ModelBlock->Block_List[j].Max_Lag;
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size_exo;i++)
{
int eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_X_Index[i];
int eqr=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_X[i];
int varr=ModelBlock->Block_List[j].IM_lead_lag[m].Exogenous[i];
int var=ModelBlock->Block_List[j].IM_lead_lag[m].Exogenous_Index[i];
output << " g1_x(" << eqr+1 << ", "
<< jacobian_max_endo_col+(m-(ModelBlock->Block_List[j].Max_Lag-ModelBlock->Block_List[j].Max_Lag_Exo))*ModelBlock->Block_List[j].nb_exo+varr+1 << ") = ";
writeDerivative(output, eq, symbol_table.getID(eExogenous, var), k, oMatlabDynamicModelSparse, temporary_terms);
output << "; % variable (exogenous)=" << symbol_table.getName(var)
<< "(" << k << ") " << var+1 << " " << varr+1
<< ", equation=" << eq+1 << endl;
}
}*/
for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
k=m-ModelBlock->Block_List[j].Max_Lag;
if (block_triangular.incidencematrix.Model_Max_Lag_Endo - ModelBlock->Block_List[j].Max_Lag +m >=0)
{
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size_other_endo;i++)
{
int eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_Index_other_endo[i];
int var=ModelBlock->Block_List[j].IM_lead_lag[m].Var_Index_other_endo[i];
int eqr=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_other_endo[i];
int varr=ModelBlock->Block_List[j].IM_lead_lag[m].Var_other_endo[i];
output << " g1_o(" << eqr+1 << ", "
<< varr+1+(m+max_endo_lag-ModelBlock->Block_List[j].Max_Lag)*symbol_table.endo_nbr() << ") = ";
writeDerivative(output, eq, symbol_table.getID(eEndogenous, var), k, oMatlabDynamicModelSparse, temporary_terms);
output << "; % variable=" << symbol_table.getName(symbol_table.getID(eEndogenous, var))
<< "(" << k << ") " << var+1
<< ", equation=" << eq+1 << endl;
}
}
}
output << " varargout{1}=g1_x;\n";
output << " varargout{2}=g1_o;\n";
output << " end;\n";
//output << " ya = y;\n";
output << " end;\n";
break;
default: break;
}
output.close();
}
}
void
DynamicModel::writeModelEquationsCodeOrdered(const string file_name, const Model_Block *ModelBlock, const string bin_basename, map_idx_type map_idx) const
{
struct Uff_l
{
int u, var, lag;
Uff_l *pNext;
};
struct Uff
{
Uff_l *Ufl, *Ufl_First;
};
int i,j,k,v;
string tmp_s;
ostringstream tmp_output;
ofstream code_file;
NodeID lhs=NULL, rhs=NULL;
BinaryOpNode *eq_node;
Uff Uf[symbol_table.endo_nbr()];
map<NodeID, int> reference_count;
vector<int> feedback_variables;
bool file_open=false;
string main_name=file_name;
main_name+=".cod";
code_file.open(main_name.c_str(), ios::out | ios::binary | ios::ate );
if (!code_file.is_open())
{
cout << "Error : Can't open file \"" << main_name << "\" for writing\n";
exit(EXIT_FAILURE);
}
//Temporary variables declaration
code_file.write(&FDIMT, sizeof(FDIMT));
k=temporary_terms.size();
code_file.write(reinterpret_cast<char *>(&k),sizeof(k));
for (j = 0; j < ModelBlock->Size ;j++)
{
feedback_variables.clear();
if (j>0)
code_file.write(&FENDBLOCK, sizeof(FENDBLOCK));
code_file.write(&FBEGINBLOCK, sizeof(FBEGINBLOCK));
v=ModelBlock->Block_List[j].Size - ModelBlock->Block_List[j].Nb_Recursives;
//cout << "v (Size) = " << v << "\n";
code_file.write(reinterpret_cast<char *>(&v),sizeof(v));
v=ModelBlock->Block_List[j].Simulation_Type;
code_file.write(reinterpret_cast<char *>(&v),sizeof(v));
int count_u;
for (i=ModelBlock->Block_List[j].Nb_Recursives; i < ModelBlock->Block_List[j].Size;i++)
{
code_file.write(reinterpret_cast<char *>(&ModelBlock->Block_List[j].Variable[i]),sizeof(ModelBlock->Block_List[j].Variable[i]));
code_file.write(reinterpret_cast<char *>(&ModelBlock->Block_List[j].Equation[i]),sizeof(ModelBlock->Block_List[j].Equation[i]));
code_file.write(reinterpret_cast<char *>(&ModelBlock->Block_List[j].Own_Derivative[i]),sizeof(ModelBlock->Block_List[j].Own_Derivative[i]));
}
if (ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_SIMPLE || ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_COMPLETE ||
ModelBlock->Block_List[j].Simulation_Type==SOLVE_BACKWARD_COMPLETE || ModelBlock->Block_List[j].Simulation_Type==SOLVE_FORWARD_COMPLETE)
{
int u_count_int=0;
//cout << "ModelBlock->Block_List[j].Nb_Recursives = " << ModelBlock->Block_List[j].Nb_Recursives << "\n";
Write_Inf_To_Bin_File(file_name, bin_basename, j, u_count_int,file_open,
ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_COMPLETE || ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_SIMPLE);
//cout << "u_count_int=" << u_count_int << "\n";
code_file.write(reinterpret_cast<char *>(&ModelBlock->Block_List[j].is_linear),sizeof(ModelBlock->Block_List[j].is_linear)); //v=block_triangular.ModelBlock->Block_List[j].IM_lead_lag[block_triangular.ModelBlock->Block_List[j].Max_Lag + block_triangular.ModelBlock->Block_List[j].Max_Lead].u_finish + 1;
v = u_count_int ;
code_file.write(reinterpret_cast<char *>(&v),sizeof(v));
v=symbol_table.endo_nbr();
code_file.write(reinterpret_cast<char *>(&v),sizeof(v));
v=block_triangular.ModelBlock->Block_List[j].Max_Lag;
code_file.write(reinterpret_cast<char *>(&v),sizeof(v));
v=block_triangular.ModelBlock->Block_List[j].Max_Lead;
code_file.write(reinterpret_cast<char *>(&v),sizeof(v));
v=u_count_int;
code_file.write(reinterpret_cast<char *>(&v),sizeof(v));
file_open=true;
}
// The equations
for (i = 0;i < ModelBlock->Block_List[j].Size;i++)
{
//The Temporary terms
temporary_terms_type tt2;
tt2.clear();
#ifdef DEBUGC
k=0;
#endif
for (temporary_terms_type::const_iterator it = ModelBlock->Block_List[j].Temporary_Terms_in_Equation[i]->begin();
it != ModelBlock->Block_List[j].Temporary_Terms_in_Equation[i]->end(); it++)
{
(*it)->compile(code_file, false, tt2, map_idx, true, false);
code_file.write(&FSTPT, sizeof(FSTPT));
map_idx_type::const_iterator ii=map_idx.find((*it)->idx);
v=(int)ii->second;
code_file.write(reinterpret_cast<char *>(&v), sizeof(v));
// Insert current node into tt2
tt2.insert(*it);
#ifdef DEBUGC
cout << "FSTPT " << v << "\n";
code_file.write(&FOK, sizeof(FOK));
code_file.write(reinterpret_cast<char *>(&k), sizeof(k));
ki++;
#endif
}
#ifdef DEBUGC
for (temporary_terms_type::const_iterator it = ModelBlock->Block_List[j].Temporary_terms->begin();
it != ModelBlock->Block_List[j].Temporary_terms->end(); it++)
{
map_idx_type::const_iterator ii=map_idx.find((*it)->idx);
cout << "map_idx[" << (*it)->idx <<"]=" << ii->second << "\n";
}
#endif
switch (ModelBlock->Block_List[j].Simulation_Type)
{
evaluation:
case EVALUATE_BACKWARD:
case EVALUATE_FORWARD:
if (ModelBlock->Block_List[j].Equation_Type[i] == E_EVALUATE)
{
eq_node = equations[ModelBlock->Block_List[j].Equation[i]];
lhs = eq_node->get_arg1();
rhs = eq_node->get_arg2();
rhs->compile(code_file, false, temporary_terms, map_idx, true, false);
lhs->compile(code_file, true, temporary_terms, map_idx, true, false);
}
else if (ModelBlock->Block_List[j].Equation_Type[i] == E_EVALUATE_S)
{
eq_node = (BinaryOpNode*)ModelBlock->Block_List[j].Equation_Normalized[i];
lhs = eq_node->get_arg1();
rhs = eq_node->get_arg2();
rhs->compile(code_file, false, temporary_terms, map_idx, true, false);
lhs->compile(code_file, true, temporary_terms, map_idx, true, false);
} break;
case SOLVE_BACKWARD_COMPLETE:
case SOLVE_FORWARD_COMPLETE:
case SOLVE_TWO_BOUNDARIES_COMPLETE:
case SOLVE_TWO_BOUNDARIES_SIMPLE:
if (i<ModelBlock->Block_List[j].Nb_Recursives)
goto evaluation;
feedback_variables.push_back(ModelBlock->Block_List[j].Variable[i]);
v=ModelBlock->Block_List[j].Equation[i];
Uf[v].Ufl=NULL;
goto end;
default:
end:
eq_node = equations[ModelBlock->Block_List[j].Equation[i]];
lhs = eq_node->get_arg1();
rhs = eq_node->get_arg2();
lhs->compile(code_file, false, temporary_terms, map_idx, true, false);
rhs->compile(code_file, false, temporary_terms, map_idx, true, false);
code_file.write(&FBINARY, sizeof(FBINARY));
int v=oMinus;
code_file.write(reinterpret_cast<char *>(&v),sizeof(v));
code_file.write(&FSTPR, sizeof(FSTPR));
v = i - ModelBlock->Block_List[j].Nb_Recursives;
code_file.write(reinterpret_cast<char *>(&v), sizeof(v));
}
}
code_file.write(&FENDEQU, sizeof(FENDEQU));
// The Jacobian if we have to solve the block
if (ModelBlock->Block_List[j].Simulation_Type!=EVALUATE_BACKWARD
&& ModelBlock->Block_List[j].Simulation_Type!=EVALUATE_FORWARD)
{
switch (ModelBlock->Block_List[j].Simulation_Type)
{
case SOLVE_BACKWARD_SIMPLE:
case SOLVE_FORWARD_SIMPLE:
compileDerivative(code_file, ModelBlock->Block_List[j].Equation[0], ModelBlock->Block_List[j].Variable[0], 0, map_idx);
code_file.write(&FSTPG, sizeof(FSTPG));
v=0;
code_file.write(reinterpret_cast<char *>(&v), sizeof(v));
break;
case SOLVE_BACKWARD_COMPLETE:
case SOLVE_FORWARD_COMPLETE:
case SOLVE_TWO_BOUNDARIES_COMPLETE:
case SOLVE_TWO_BOUNDARIES_SIMPLE:
//count_u=ModelBlock->Block_List[j].Size - ModelBlock->Block_List[j].Nb_Recursives;
count_u = feedback_variables.size();
for(i=0; i<(int)ModelBlock->Block_List[j].Chain_Rule_Derivatives->size();i++)
{
pair< pair<int, pair<int, int> >, pair<int, int> > it = ModelBlock->Block_List[j].Chain_Rule_Derivatives->at(i);
k=it.first.first;
int eq=it.first.second.first;
int var=it.first.second.second;
int eqr=it.second.first;
int varr=it.second.second;
//cout << "k=" << k << " eq=" << eq << " (" << eq-ModelBlock->Block_List[j].Nb_Recursives << ") var=" << var << " (" << var-ModelBlock->Block_List[j].Nb_Recursives << ") eqr=" << eqr << " varr=" << varr << " count_u=" << count_u << "\n";
int v=ModelBlock->Block_List[j].Equation[eq];
/*m = ModelBlock->Block_List[j].Max_Lag + k;
int u=ModelBlock->Block_List[j].IM_lead_lag[m].u[i];*/
if(eq>=ModelBlock->Block_List[j].Nb_Recursives and var>=ModelBlock->Block_List[j].Nb_Recursives)
{
if (!Uf[v].Ufl)
{
Uf[v].Ufl=(Uff_l*)malloc(sizeof(Uff_l));
Uf[v].Ufl_First=Uf[v].Ufl;
}
else
{
Uf[v].Ufl->pNext=(Uff_l*)malloc(sizeof(Uff_l));
Uf[v].Ufl=Uf[v].Ufl->pNext; }
Uf[v].Ufl->pNext=NULL;
Uf[v].Ufl->u=count_u;
Uf[v].Ufl->var=varr;
Uf[v].Ufl->lag=k;
compileChainRuleDerivative(code_file, eqr, varr, k, map_idx);
code_file.write(&FSTPU, sizeof(FSTPU));
code_file.write(reinterpret_cast<char *>(&count_u), sizeof(count_u));
count_u++;
}
}
for (i = 0;i < ModelBlock->Block_List[j].Size;i++)
{
if(i>=ModelBlock->Block_List[j].Nb_Recursives)
{
code_file.write(&FLDR, sizeof(FLDR));
v = i-ModelBlock->Block_List[j].Nb_Recursives;
code_file.write(reinterpret_cast<char *>(&v), sizeof(v));
code_file.write(&FLDZ, sizeof(FLDZ));
v=ModelBlock->Block_List[j].Equation[i];
for (Uf[v].Ufl=Uf[v].Ufl_First; Uf[v].Ufl; Uf[v].Ufl=Uf[v].Ufl->pNext)
{
code_file.write(&FLDU, sizeof(FLDU));
code_file.write(reinterpret_cast<char *>(&Uf[v].Ufl->u), sizeof(Uf[v].Ufl->u));
code_file.write(&FLDV, sizeof(FLDV));
char vc=eEndogenous;
code_file.write(reinterpret_cast<char *>(&vc), sizeof(vc));
int v1=Uf[v].Ufl->var;
code_file.write(reinterpret_cast<char *>(&v1), sizeof(v1));
v1=Uf[v].Ufl->lag;
code_file.write(reinterpret_cast<char *>(&v1), sizeof(v1));
code_file.write(&FBINARY, sizeof(FBINARY));
v1=oTimes;
code_file.write(reinterpret_cast<char *>(&v1), sizeof(v1));
code_file.write(&FCUML, sizeof(FCUML));
}
Uf[v].Ufl=Uf[v].Ufl_First;
while (Uf[v].Ufl)
{
Uf[v].Ufl_First=Uf[v].Ufl->pNext;
free(Uf[v].Ufl);
Uf[v].Ufl=Uf[v].Ufl_First;
}
code_file.write(&FBINARY, sizeof(FBINARY));
v=oMinus;
code_file.write(reinterpret_cast<char *>(&v), sizeof(v));
code_file.write(&FSTPU, sizeof(FSTPU));
v = i - ModelBlock->Block_List[j].Nb_Recursives;
code_file.write(reinterpret_cast<char *>(&v), sizeof(v));
}
}
break;
default:
break;
}
}
}
code_file.write(&FENDBLOCK, sizeof(FENDBLOCK));
code_file.write(&FEND, sizeof(FEND));
code_file.close();
}
void
DynamicModel::writeDynamicMFile(const string &dynamic_basename) const
{
string filename = dynamic_basename + ".m";
ofstream mDynamicModelFile;
mDynamicModelFile.open(filename.c_str(), ios::out | ios::binary);
if (!mDynamicModelFile.is_open()) {
cerr << "Error: Can't open file " << filename << " for writing" << endl;
exit(EXIT_FAILURE);
}
mDynamicModelFile << "function [residual, g1, g2, g3] = " << dynamic_basename << "(y, x, params, it_)" << endl
<< "%" << endl
<< "% Status : Computes dynamic model for Dynare" << endl
<< "%" << endl
<< "% Warning : this file is generated automatically by Dynare" << endl
<< "% from model file (.mod)" << endl << endl;
writeDynamicModel(mDynamicModelFile, false);
mDynamicModelFile.close();
}
void
DynamicModel::writeDynamicCFile(const string &dynamic_basename) const
{
string filename = dynamic_basename + ".c";
ofstream mDynamicModelFile;
mDynamicModelFile.open(filename.c_str(), ios::out | ios::binary);
if (!mDynamicModelFile.is_open())
{
cerr << "Error: Can't open file " << filename << " for writing" << endl;
exit(EXIT_FAILURE);
}
mDynamicModelFile << "/*" << endl
<< " * " << filename << " : Computes dynamic model for Dynare" << endl
<< " *" << endl
<< " * Warning : this file is generated automatically by Dynare" << endl
<< " * from model file (.mod)" << endl
<< endl
<< " */" << endl
<< "#include <math.h>" << endl
<< "#include \"mex.h\"" << endl;
// Writing the function body
writeDynamicModel(mDynamicModelFile, true);
// Writing the gateway routine
mDynamicModelFile << "/* The gateway routine */" << endl
<< "void mexFunction(int nlhs, mxArray *plhs[], int nrhs, const mxArray *prhs[])" << endl
<< "{" << endl
<< " double *y, *x, *params;" << endl
<< " double *residual, *g1, *v2;" << endl
<< " int nb_row_x, it_;" << endl
<< endl
<< " /* Create a pointer to the input matrix y. */" << endl
<< " y = mxGetPr(prhs[0]);" << endl
<< endl
<< " /* Create a pointer to the input matrix x. */" << endl
<< " x = mxGetPr(prhs[1]);" << endl
<< endl
<< " /* Create a pointer to the input matrix params. */" << endl
<< " params = mxGetPr(prhs[2]);" << endl
<< endl
<< " /* Fetch time index */" << endl
<< " it_ = (int) mxGetScalar(prhs[3]) - 1;" << endl
<< endl
<< " /* Gets number of rows of matrix x. */" << endl
<< " nb_row_x = mxGetM(prhs[1]);" << endl
<< endl
<< " residual = NULL;" << endl
<< " if (nlhs >= 1)" << endl
<< " {" << endl
<< " /* Set the output pointer to the output matrix residual. */" << endl
<< " plhs[0] = mxCreateDoubleMatrix(" << equations.size() << ",1, mxREAL);" << endl
<< " /* Create a C pointer to a copy of the output matrix residual. */" << endl << " residual = mxGetPr(plhs[0]);" << endl
<< " }" << endl
<< endl
<< " g1 = NULL;" << endl
<< " if (nlhs >= 2)" << endl
<< " {" << endl
<< " /* Set the output pointer to the output matrix g1. */" << endl
<< " plhs[1] = mxCreateDoubleMatrix(" << equations.size() << ", " << dynJacobianColsNbr << ", mxREAL);" << endl
<< " /* Create a C pointer to a copy of the output matrix g1. */" << endl
<< " g1 = mxGetPr(plhs[1]);" << endl
<< " }" << endl
<< endl
<< " v2 = NULL;" << endl
<< " if (nlhs >= 3)" << endl
<< " {" << endl
<< " /* Set the output pointer to the output matrix v2. */" << endl
<< " plhs[2] = mxCreateDoubleMatrix(" << NNZDerivatives[1] << ", " << 3
<< ", mxREAL);" << endl
<< " /* Create a C pointer to a copy of the output matrix g1. */" << endl
<< " v2 = mxGetPr(plhs[2]);" << endl
<< " }" << endl
<< endl
<< " /* Call the C subroutines. */" << endl
<< " Dynamic(y, x, nb_row_x, params, it_, residual, g1, v2);" << endl
<< "}" << endl;
mDynamicModelFile.close();
}
string
DynamicModel::reform(const string name1) const
{
string name=name1;
int pos = name.find("\\", 0);
while (pos >= 0)
{
if (name.substr(pos + 1, 1) != "\\")
{
name = name.insert(pos, "\\");
pos++;
}
pos++;
pos = name.find("\\", pos);
}
return (name);
}
void
DynamicModel::Write_Inf_To_Bin_File(const string &dynamic_basename, const string &bin_basename, const int &num,
int &u_count_int, bool &file_open, bool is_two_boundaries) const
{
int j;
std::ofstream SaveCode;
if (file_open)
SaveCode.open((bin_basename + "_dynamic.bin").c_str(), ios::out | ios::in | ios::binary | ios ::ate );
else
SaveCode.open((bin_basename + "_dynamic.bin").c_str(), ios::out | ios::binary);
if (!SaveCode.is_open())
{
cout << "Error : Can't open file \"" << bin_basename << "_dynamic.bin\" for writing\n";
exit(EXIT_FAILURE);
}
u_count_int=0;
int Size = block_triangular.ModelBlock->Block_List[num].Size - block_triangular.ModelBlock->Block_List[num].Nb_Recursives;
for(int i=0; i<(int)block_triangular.ModelBlock->Block_List[num].Chain_Rule_Derivatives->size();i++)
{
//Chain_Rule_Derivatives.insert(make_pair( make_pair(eq, eqr), make_pair(var, make_pair(varr, lag))));
pair< pair<int, pair<int, int> >, pair<int, int> > it = block_triangular.ModelBlock->Block_List[num].Chain_Rule_Derivatives->at(i);
int k=it.first.first;
int eq=it.first.second.first;
int var_init=it.first.second.second;
/*int eqr=it.second.first;
int varr=it.second.second;*/
if(eq>=block_triangular.ModelBlock->Block_List[num].Nb_Recursives and var_init>=block_triangular.ModelBlock->Block_List[num].Nb_Recursives)
{
int v=eq-block_triangular.ModelBlock->Block_List[num].Nb_Recursives;
SaveCode.write(reinterpret_cast<char *>(&v), sizeof(v));
int var=it.first.second.second-block_triangular.ModelBlock->Block_List[num].Nb_Recursives + k * Size;
SaveCode.write(reinterpret_cast<char *>(&var), sizeof(var));
SaveCode.write(reinterpret_cast<char *>(&k), sizeof(k));
int u = u_count_int + Size;
SaveCode.write(reinterpret_cast<char *>(&u), sizeof(u));
//cout << "eq=" << eq << " var=" << var << " k=" << k << " u=" << u << "\n";
u_count_int++;
}
}
/*for (int m=0;m<=block_triangular.ModelBlock->Block_List[num].Max_Lead+block_triangular.ModelBlock->Block_List[num].Max_Lag;m++)
{
int k1=m-block_triangular.ModelBlock->Block_List[num].Max_Lag;
for (j=0;j<block_triangular.ModelBlock->Block_List[num].IM_lead_lag[m].size;j++)
{
int varr=block_triangular.ModelBlock->Block_List[num].IM_lead_lag[m].Var[j]+k1*block_triangular.ModelBlock->Block_List[num].Size;
int u=block_triangular.ModelBlock->Block_List[num].IM_lead_lag[m].u[j];
int eqr1=block_triangular.ModelBlock->Block_List[num].IM_lead_lag[m].Equ[j];
SaveCode.write(reinterpret_cast<char *>(&eqr1), sizeof(eqr1));
SaveCode.write(reinterpret_cast<char *>(&varr), sizeof(varr));
SaveCode.write(reinterpret_cast<char *>(&k1), sizeof(k1));
SaveCode.write(reinterpret_cast<char *>(&u), sizeof(u));
u_count_int++;
}
}*/
if (is_two_boundaries)
{
/*for (j=0;j<Size;j++)
{
int eqr1=j;
int varr=Size*(block_triangular.periods
+block_triangular.incidencematrix.Model_Max_Lead_Endo);
int k1=0;
SaveCode.write(reinterpret_cast<char *>(&eqr1), sizeof(eqr1));
SaveCode.write(reinterpret_cast<char *>(&varr), sizeof(varr));
SaveCode.write(reinterpret_cast<char *>(&k1), sizeof(k1));
SaveCode.write(reinterpret_cast<char *>(&eqr1), sizeof(eqr1));
u_count_int++;
}*/
u_count_int+=Size;
}
//cout << "u_count_int=" << u_count_int << "\n";
for (j=block_triangular.ModelBlock->Block_List[num].Nb_Recursives;j<block_triangular.ModelBlock->Block_List[num].Size;j++)
{
int varr=block_triangular.ModelBlock->Block_List[num].Variable[j];
SaveCode.write(reinterpret_cast<char *>(&varr), sizeof(varr));
}
for (j=block_triangular.ModelBlock->Block_List[num].Nb_Recursives;j<block_triangular.ModelBlock->Block_List[num].Size;j++)
{
int eqr1=block_triangular.ModelBlock->Block_List[num].Equation[j];
SaveCode.write(reinterpret_cast<char *>(&eqr1), sizeof(eqr1));
}
SaveCode.close();
}
void
DynamicModel::writeSparseDynamicMFile(const string &dynamic_basename, const string &basename) const
{
string sp;
ofstream mDynamicModelFile;
ostringstream tmp, tmp1, tmp_eq; int prev_Simulation_Type, tmp_i;
//SymbolicGaussElimination SGE;
bool OK;
chdir(basename.c_str());
string filename = dynamic_basename + ".m";
mDynamicModelFile.open(filename.c_str(), ios::out | ios::binary);
if (!mDynamicModelFile.is_open())
{
cerr << "Error: Can't open file " << filename << " for writing" << endl;
exit(EXIT_FAILURE);
}
mDynamicModelFile << "%\n";
mDynamicModelFile << "% " << filename << " : Computes dynamic model for Dynare\n";
mDynamicModelFile << "%\n";
mDynamicModelFile << "% Warning : this file is generated automatically by Dynare\n";
mDynamicModelFile << "% from model file (.mod)\n\n";
mDynamicModelFile << "%/\n";
int i, k, Nb_SGE=0;
bool skip_head, open_par=false;
mDynamicModelFile << "function [varargout] = " << dynamic_basename << "(varargin)\n";
mDynamicModelFile << " global oo_ options_ M_ ;\n";
mDynamicModelFile << " g2=[];g3=[];\n";
//Temporary variables declaration
OK=true;
ostringstream tmp_output;
for (temporary_terms_type::const_iterator it = temporary_terms.begin();
it != temporary_terms.end(); it++)
{
if (OK)
OK=false;
else
tmp_output << " ";
(*it)->writeOutput(tmp_output, oMatlabStaticModelSparse, temporary_terms);
}
if (tmp_output.str().length()>0)
mDynamicModelFile << " global " << tmp_output.str() << " M_ ;\n";
mDynamicModelFile << " T_init=zeros(1,options_.periods+M_.maximum_lag+M_.maximum_lead);\n";
tmp_output.str("");
for (temporary_terms_type::const_iterator it = temporary_terms.begin();
it != temporary_terms.end(); it++)
{
tmp_output << " ";
(*it)->writeOutput(tmp_output, oMatlabDynamicModel, temporary_terms);
tmp_output << "=T_init;\n";
}
if (tmp_output.str().length()>0)
mDynamicModelFile << tmp_output.str();
mDynamicModelFile << " y_kmin=M_.maximum_lag;\n";
mDynamicModelFile << " y_kmax=M_.maximum_lead;\n";
mDynamicModelFile << " y_size=M_.endo_nbr;\n";
mDynamicModelFile << " if(length(varargin)>0)\n";
mDynamicModelFile << " %it is a simple evaluation of the dynamic model for time _it\n";
mDynamicModelFile << " params=varargin{3};\n";
mDynamicModelFile << " it_=varargin{4};\n";
/*i = symbol_table.endo_nbr*(variable_table.max_endo_lag+variable_table.max_endo_lead+1)+
symbol_table.exo_nbr*(variable_table.max_exo_lag+variable_table.max_exo_lead+1);
mDynamicModelFile << " g1=spalloc(" << symbol_table.endo_nbr << ", " << i << ", " << i*symbol_table.endo_nbr << ");\n";*/
mDynamicModelFile << " Per_u_=0;\n";
mDynamicModelFile << " Per_y_=it_*y_size;\n";
mDynamicModelFile << " y=varargin{1};\n";
mDynamicModelFile << " ys=y(it_,:);\n";
mDynamicModelFile << " x=varargin{2};\n";
prev_Simulation_Type=-1;
tmp.str("");
tmp_eq.str("");
for (int count_call=1, i = 0;i < block_triangular.ModelBlock->Size;i++, count_call++) {
mDynamicModelFile << " %block_triangular.ModelBlock->Block_List[i].Nb_Recursives=" << block_triangular.ModelBlock->Block_List[i].Nb_Recursives << " block_triangular.ModelBlock->Block_List[i].Size=" << block_triangular.ModelBlock->Block_List[i].Size << "\n";
k=block_triangular.ModelBlock->Block_List[i].Simulation_Type;
if(k==EVALUATE_FORWARD || k==EVALUATE_BACKWARD)
{
for (int ik=0 ;ik<block_triangular.ModelBlock->Block_List[i].Size;ik++)
{
tmp << " " << block_triangular.ModelBlock->Block_List[i].Variable[ik]+1;
tmp_eq << " " << block_triangular.ModelBlock->Block_List[i].Equation[ik]+1;
}
}
else
{
for (int ik=block_triangular.ModelBlock->Block_List[i].Nb_Recursives ;ik<block_triangular.ModelBlock->Block_List[i].Size;ik++)
{
tmp << " " << block_triangular.ModelBlock->Block_List[i].Variable[ik]+1;
tmp_eq << " " << block_triangular.ModelBlock->Block_List[i].Equation[ik]+1;
}
}
mDynamicModelFile << " y_index_eq=[" << tmp_eq.str() << "];\n";
mDynamicModelFile << " y_index=[" << tmp.str() << "];\n";
switch (k)
{
case EVALUATE_FORWARD:
case EVALUATE_BACKWARD:
mDynamicModelFile << " [y, dr(" << count_call << ").g1, dr(" << count_call << ").g2, dr(" << count_call << ").g3, dr(" << count_call << ").g1_x, dr(" << count_call << ").g1_o]=" << dynamic_basename << "_" << i + 1 << "(y, x, params, 1, it_-1, 1);\n";
mDynamicModelFile << " residual(y_index_eq)=ys(y_index)-y(it_, y_index);\n";
break;
case SOLVE_FORWARD_SIMPLE:
case SOLVE_BACKWARD_SIMPLE:
//mDynamicModelFile << " y_index_eq = " << block_triangular.ModelBlock->Block_List[i].Equation[0]+1 << ";\n";
mDynamicModelFile << " [r, dr(" << count_call << ").g1, dr(" << count_call << ").g2, dr(" << count_call << ").g3, dr(" << count_call << ").g1_x, dr(" << count_call << ").g1_o]=" << dynamic_basename << "_" << i + 1 << "(y, x, params, it_, 1);\n";
mDynamicModelFile << " residual(y_index_eq)=r;\n";
break;
case SOLVE_FORWARD_COMPLETE:
case SOLVE_BACKWARD_COMPLETE:
//mDynamicModelFile << " y_index_eq = [" << tmp_eq.str() << "];\n";
mDynamicModelFile << " [r, dr(" << count_call << ").g1, dr(" << count_call << ").g2, dr(" << count_call << ").g3, dr(" << count_call << ").g1_x, dr(" << count_call << ").g1_o]=" << dynamic_basename << "_" << i + 1 << "(y, x, params, it_, 1);\n";
mDynamicModelFile << " residual(y_index_eq)=r;\n";
break;
case SOLVE_TWO_BOUNDARIES_COMPLETE:
case SOLVE_TWO_BOUNDARIES_SIMPLE:
int j;
/*mDynamicModelFile << " y_index_eq = [" << tmp_eq.str() << "];\n";
tmp_i=block_triangular.ModelBlock->Block_List[i].Max_Lag_Endo+block_triangular.ModelBlock->Block_List[i].Max_Lead_Endo+1;
mDynamicModelFile << " y_index = [";
for (j=0;j<tmp_i;j++)
for (int ik=0;ik<block_triangular.ModelBlock->Block_List[i].Size;ik++)
{
mDynamicModelFile << " " << block_triangular.ModelBlock->Block_List[i].Variable[ik]+1+j*symbol_table.endo_nbr();
}
int tmp_ix=block_triangular.ModelBlock->Block_List[i].Max_Lag_Exo+block_triangular.ModelBlock->Block_List[i].Max_Lead_Exo+1;
for (j=0;j<tmp_ix;j++)
for (int ik=0;ik<block_triangular.ModelBlock->Block_List[i].nb_exo;ik++)
mDynamicModelFile << " " << block_triangular.ModelBlock->Block_List[i].Exogenous[ik]+1+j*symbol_table.exo_nbr()+symbol_table.endo_nbr()*tmp_i;
mDynamicModelFile << " ];\n";*/
//mDynamicModelFile << " ga = [];\n";
j = block_triangular.ModelBlock->Block_List[i].Size*(block_triangular.ModelBlock->Block_List[i].Max_Lag_Endo+block_triangular.ModelBlock->Block_List[i].Max_Lead_Endo+1)
+ block_triangular.ModelBlock->Block_List[i].nb_exo*(block_triangular.ModelBlock->Block_List[i].Max_Lag_Exo+block_triangular.ModelBlock->Block_List[i].Max_Lead_Exo+1);
/*mDynamicModelFile << " ga=spalloc(" << block_triangular.ModelBlock->Block_List[i].Size << ", " << j << ", " <<
block_triangular.ModelBlock->Block_List[i].Size*j << ");\n";*/
tmp_i=block_triangular.ModelBlock->Block_List[i].Max_Lag_Endo+block_triangular.ModelBlock->Block_List[i].Max_Lead_Endo+1;
mDynamicModelFile << " [r, dr(" << count_call << ").g1, dr(" << count_call << ").g2, dr(" << count_call << ").g3, b, dr(" << count_call << ").g1_x, dr(" << count_call << ").g1_o]=" << dynamic_basename << "_" << i + 1 << "(y, x, params, it_-" << max_lag << ", 1, " << max_lag << ", " << block_triangular.ModelBlock->Block_List[i].Size-block_triangular.ModelBlock->Block_List[i].Nb_Recursives << ");\n";
/*if(block_triangular.ModelBlock->Block_List[i].Max_Lag==variable_table.max_lag && block_triangular.ModelBlock->Block_List[i].Max_Lead==variable_table.max_lead)
mDynamicModelFile << " g1(y_index_eq,y_index) = ga;\n";
else
mDynamicModelFile << " g1(y_index_eq,y_index) = ga(:," << 1+(variable_table.max_lag-block_triangular.ModelBlock->Block_List[i].Max_Lag)*block_triangular.ModelBlock->Block_List[i].Size << ":" << (variable_table.max_lag+1+block_triangular.ModelBlock->Block_List[i].Max_Lead)*block_triangular.ModelBlock->Block_List[i].Size << ");\n";*/
mDynamicModelFile << " residual(y_index_eq)=r(:,M_.maximum_lag+1);\n";
break; }
tmp_eq.str("");
tmp.str("");
}
if (tmp1.str().length())
{
mDynamicModelFile << tmp1.str();
tmp1.str("");
}
mDynamicModelFile << " varargout{1}=residual;\n";
mDynamicModelFile << " varargout{2}=dr;\n";
mDynamicModelFile << " return;\n";
mDynamicModelFile << " end;\n";
mDynamicModelFile << " %it is the deterministic simulation of the block decomposed dynamic model\n";
mDynamicModelFile << " if(options_.stack_solve_algo==1)\n";
mDynamicModelFile << " mthd='Sparse LU';\n";
mDynamicModelFile << " elseif(options_.stack_solve_algo==2)\n";
mDynamicModelFile << " mthd='GMRES';\n";
mDynamicModelFile << " elseif(options_.stack_solve_algo==3)\n";
mDynamicModelFile << " mthd='BICGSTAB';\n";
mDynamicModelFile << " elseif(options_.stack_solve_algo==4)\n";
mDynamicModelFile << " mthd='OPTIMPATH';\n";
mDynamicModelFile << " else\n";
mDynamicModelFile << " mthd='UNKNOWN';\n";
mDynamicModelFile << " end;\n";
mDynamicModelFile << " disp (['-----------------------------------------------------']) ;\n";
mDynamicModelFile << " disp (['MODEL SIMULATION: (method=' mthd ')']) ;\n";
mDynamicModelFile << " fprintf('\\n') ;\n";
mDynamicModelFile << " periods=options_.periods;\n";
mDynamicModelFile << " maxit_=options_.maxit_;\n";
mDynamicModelFile << " solve_tolf=options_.solve_tolf;\n";
mDynamicModelFile << " y=oo_.endo_simul';\n";
mDynamicModelFile << " x=oo_.exo_simul;\n";
prev_Simulation_Type=-1;
mDynamicModelFile << " params=M_.params;\n";
mDynamicModelFile << " oo_.deterministic_simulation.status = 0;\n";
for (i = 0;i < block_triangular.ModelBlock->Size;i++)
{
k = block_triangular.ModelBlock->Block_List[i].Simulation_Type;
if (BlockTriangular::BlockSim(prev_Simulation_Type)==BlockTriangular::BlockSim(k) &&
(k==EVALUATE_FORWARD || k==EVALUATE_BACKWARD /*|| k==EVALUATE_FORWARD_R || k==EVALUATE_BACKWARD_R*/))
skip_head=true;
else
skip_head=false;
if ((k == EVALUATE_FORWARD /*|| k == EVALUATE_FORWARD_R*/) && (block_triangular.ModelBlock->Block_List[i].Size))
{
if (!skip_head)
{
if (open_par)
{
mDynamicModelFile << " end\n";
}
mDynamicModelFile << " oo_.deterministic_simulation.status = 1;\n";
mDynamicModelFile << " oo_.deterministic_simulation.error = 0;\n";
mDynamicModelFile << " oo_.deterministic_simulation.iterations = 0;\n";
mDynamicModelFile << " if(isfield(oo_.deterministic_simulation,'block'))\n";
mDynamicModelFile << " blck_num = length(oo_.deterministic_simulation.block)+1;\n";
mDynamicModelFile << " else\n";
mDynamicModelFile << " blck_num = 1;\n";
mDynamicModelFile << " end;\n";
mDynamicModelFile << " oo_.deterministic_simulation.block(blck_num).status = 1;\n";
mDynamicModelFile << " oo_.deterministic_simulation.block(blck_num).error = 0;\n";
mDynamicModelFile << " oo_.deterministic_simulation.block(blck_num).iterations = 0;\n";
mDynamicModelFile << " g1=[];g2=[];g3=[];\n";
//mDynamicModelFile << " for it_ = y_kmin+1:(periods+y_kmin)\n";
mDynamicModelFile << " y=" << dynamic_basename << "_" << i + 1 << "(y, x, params, 0, y_kmin, periods);\n";
mDynamicModelFile << " tmp = y(:,M_.block_structure.block(" << i + 1 << ").variable);\n";
mDynamicModelFile << " if(isnan(tmp) | isinf(tmp))\n";
mDynamicModelFile << " disp(['Inf or Nan value during the evaluation of block " << i <<"']);\n"; mDynamicModelFile << " return;\n";
mDynamicModelFile << " end;\n";
}
//open_par=true;
}
else if ((k == EVALUATE_BACKWARD /*|| k == EVALUATE_BACKWARD_R*/) && (block_triangular.ModelBlock->Block_List[i].Size))
{
if (!skip_head)
{
if (open_par)
{
mDynamicModelFile << " end\n";
}
mDynamicModelFile << " oo_.deterministic_simulation.status = 1;\n";
mDynamicModelFile << " oo_.deterministic_simulation.error = 0;\n";
mDynamicModelFile << " oo_.deterministic_simulation.iterations = 0;\n";
mDynamicModelFile << " if(isfield(oo_.deterministic_simulation,'block'))\n";
mDynamicModelFile << " blck_num = length(oo_.deterministic_simulation.block)+1;\n";
mDynamicModelFile << " else\n";
mDynamicModelFile << " blck_num = 1;\n";
mDynamicModelFile << " end;\n";
mDynamicModelFile << " oo_.deterministic_simulation.block(blck_num).status = 1;\n";
mDynamicModelFile << " oo_.deterministic_simulation.block(blck_num).error = 0;\n";
mDynamicModelFile << " oo_.deterministic_simulation.block(blck_num).iterations = 0;\n";
mDynamicModelFile << " g1=[];g2=[];g3=[];\n";
mDynamicModelFile << " " << dynamic_basename << "_" << i + 1 << "(y, x, params, 0, y_kmin, periods);\n";
mDynamicModelFile << " tmp = y(:,M_.block_structure.block(" << i + 1 << ").variable);\n";
mDynamicModelFile << " if(isnan(tmp) | isinf(tmp))\n";
mDynamicModelFile << " disp(['Inf or Nan value during the evaluation of block " << i <<"']);\n";
mDynamicModelFile << " return;\n";
mDynamicModelFile << " end;\n";
}
}
else if ((k == SOLVE_FORWARD_COMPLETE || k == SOLVE_FORWARD_SIMPLE) && (block_triangular.ModelBlock->Block_List[i].Size))
{
if (open_par)
mDynamicModelFile << " end\n";
open_par=false;
mDynamicModelFile << " g1=0;\n";
mDynamicModelFile << " r=0;\n";
tmp.str("");
for (int ik=block_triangular.ModelBlock->Block_List[i].Nb_Recursives ;ik<block_triangular.ModelBlock->Block_List[i].Size;ik++)
{
tmp << " " << block_triangular.ModelBlock->Block_List[i].Variable[ik]+1;
}
mDynamicModelFile << " y_index = [" << tmp.str() << "];\n";
int nze, m;
for (nze=0,m=0;m<=block_triangular.ModelBlock->Block_List[i].Max_Lead+block_triangular.ModelBlock->Block_List[i].Max_Lag;m++)
nze+=block_triangular.ModelBlock->Block_List[i].IM_lead_lag[m].size;
mDynamicModelFile << " if(isfield(oo_.deterministic_simulation,'block'))\n";
mDynamicModelFile << " blck_num = length(oo_.deterministic_simulation.block)+1;\n";
mDynamicModelFile << " else\n";
mDynamicModelFile << " blck_num = 1;\n";
mDynamicModelFile << " end;\n";
mDynamicModelFile << " y = solve_one_boundary('" << dynamic_basename << "_" << i + 1 << "'" <<
", y, x, params, y_index, " << nze <<
", options_.periods, " << block_triangular.ModelBlock->Block_List[i].is_linear <<
", blck_num, y_kmin, options_.maxit_, options_.solve_tolf, options_.slowc, " << cutoff << ", options_.stack_solve_algo, 1, 1, 0);\n";
mDynamicModelFile << " tmp = y(:,M_.block_structure.block(" << i + 1 << ").variable);\n";
mDynamicModelFile << " if(isnan(tmp) | isinf(tmp))\n";
mDynamicModelFile << " disp(['Inf or Nan value during the resolution of block " << i <<"']);\n";
mDynamicModelFile << " return;\n";
mDynamicModelFile << " end;\n";
}
else if ((k == SOLVE_BACKWARD_COMPLETE || k == SOLVE_BACKWARD_SIMPLE) && (block_triangular.ModelBlock->Block_List[i].Size))
{
if (open_par)
mDynamicModelFile << " end\n";
open_par=false;
mDynamicModelFile << " g1=0;\n"; mDynamicModelFile << " r=0;\n";
tmp.str("");
for (int ik=block_triangular.ModelBlock->Block_List[i].Nb_Recursives ;ik<block_triangular.ModelBlock->Block_List[i].Size;ik++)
{
tmp << " " << block_triangular.ModelBlock->Block_List[i].Variable[ik]+1;
}
mDynamicModelFile << " y_index = [" << tmp.str() << "];\n";
int nze, m;
for (nze=0,m=0;m<=block_triangular.ModelBlock->Block_List[i].Max_Lead+block_triangular.ModelBlock->Block_List[i].Max_Lag;m++)
nze+=block_triangular.ModelBlock->Block_List[i].IM_lead_lag[m].size;
mDynamicModelFile << " if(isfield(oo_.deterministic_simulation,'block'))\n";
mDynamicModelFile << " blck_num = length(oo_.deterministic_simulation.block)+1;\n";
mDynamicModelFile << " else\n";
mDynamicModelFile << " blck_num = 1;\n";
mDynamicModelFile << " end;\n";
mDynamicModelFile << " y = solve_one_boundary('" << dynamic_basename << "_" << i + 1 << "'" <<
", y, x, params, y_index, " << nze <<
", options_.periods, " << block_triangular.ModelBlock->Block_List[i].is_linear <<
", blck_num, y_kmin, options_.maxit_, options_.solve_tolf, options_.slowc, " << cutoff << ", options_.stack_solve_algo, 1, 1, 0);\n";
mDynamicModelFile << " tmp = y(:,M_.block_structure.block(" << i + 1 << ").variable);\n";
mDynamicModelFile << " if(isnan(tmp) | isinf(tmp))\n";
mDynamicModelFile << " disp(['Inf or Nan value during the resolution of block " << i <<"']);\n";
mDynamicModelFile << " return;\n";
mDynamicModelFile << " end;\n";
}
else if ((k == SOLVE_TWO_BOUNDARIES_COMPLETE || k == SOLVE_TWO_BOUNDARIES_SIMPLE) && (block_triangular.ModelBlock->Block_List[i].Size))
{
if (open_par)
mDynamicModelFile << " end\n";
open_par=false;
Nb_SGE++;
int nze, m;
for (nze=0,m=0;m<=block_triangular.ModelBlock->Block_List[i].Max_Lead+block_triangular.ModelBlock->Block_List[i].Max_Lag;m++)
nze+=block_triangular.ModelBlock->Block_List[i].IM_lead_lag[m].size;
mDynamicModelFile << " y_index=[";
for (int ik=block_triangular.ModelBlock->Block_List[i].Nb_Recursives ;ik<block_triangular.ModelBlock->Block_List[i].Size;ik++)
{
mDynamicModelFile << " " << block_triangular.ModelBlock->Block_List[i].Variable[ik]+1;
}
mDynamicModelFile << " ];\n";
mDynamicModelFile << " if(isfield(oo_.deterministic_simulation,'block'))\n";
mDynamicModelFile << " blck_num = length(oo_.deterministic_simulation.block)+1;\n";
mDynamicModelFile << " else\n";
mDynamicModelFile << " blck_num = 1;\n";
mDynamicModelFile << " end;\n";
mDynamicModelFile << " y = solve_two_boundaries('" << dynamic_basename << "_" << i + 1 << "'" <<
", y, x, params, y_index, " << nze <<
", options_.periods, " << block_triangular.ModelBlock->Block_List[i].Max_Lag <<
", " << block_triangular.ModelBlock->Block_List[i].Max_Lead <<
", " << block_triangular.ModelBlock->Block_List[i].is_linear <<
", blck_num, y_kmin, options_.maxit_, options_.solve_tolf, options_.slowc, " << cutoff << ", options_.stack_solve_algo);\n";
mDynamicModelFile << " tmp = y(:,M_.block_structure.block(" << i + 1 << ").variable);\n";
mDynamicModelFile << " if(isnan(tmp) | isinf(tmp))\n";
mDynamicModelFile << " disp(['Inf or Nan value during the resolution of block " << i <<"']);\n";
mDynamicModelFile << " return;\n";
mDynamicModelFile << " end;\n";
}
prev_Simulation_Type=k;
}
if (open_par)
mDynamicModelFile << " end;\n";
open_par=false;
mDynamicModelFile << " oo_.endo_simul = y';\n";
mDynamicModelFile << "return;\n";
mDynamicModelFile.close();
writeModelEquationsOrdered_M( block_triangular.ModelBlock, dynamic_basename);
chdir(".."); }
void
DynamicModel::writeDynamicModel(ostream &DynamicOutput, bool use_dll) const
{
ostringstream model_output; // Used for storing model equations
ostringstream jacobian_output; // Used for storing jacobian equations
ostringstream hessian_output; // Used for storing Hessian equations
ostringstream third_derivatives_output;
ExprNodeOutputType output_type = (use_dll ? oCDynamicModel : oMatlabDynamicModel);
writeModelLocalVariables(model_output, output_type);
writeTemporaryTerms(temporary_terms, model_output, output_type);
writeModelEquations(model_output, output_type);
int nrows = equations.size();
int hessianColsNbr = dynJacobianColsNbr * dynJacobianColsNbr;
// Writing Jacobian
for (first_derivatives_type::const_iterator it = first_derivatives.begin();
it != first_derivatives.end(); it++)
{
int eq = it->first.first;
int var = it->first.second;
NodeID d1 = it->second;
ostringstream g1;
g1 << " g1";
jacobianHelper(g1, eq, getDynJacobianCol(var), output_type);
jacobian_output << g1.str() << "=" << g1.str() << "+";
d1->writeOutput(jacobian_output, output_type, temporary_terms);
jacobian_output << ";" << endl;
}
// Writing Hessian
int k = 0; // Keep the line of a 2nd derivative in v2
for (second_derivatives_type::const_iterator it = second_derivatives.begin();
it != second_derivatives.end(); it++)
{
int eq = it->first.first;
int var1 = it->first.second.first;
int var2 = it->first.second.second;
NodeID d2 = it->second;
int id1 = getDynJacobianCol(var1);
int id2 = getDynJacobianCol(var2);
int col_nb = id1 * dynJacobianColsNbr + id2;
int col_nb_sym = id2 * dynJacobianColsNbr + id1;
hessian_output << "v2";
hessianHelper(hessian_output, k, 0, output_type);
hessian_output << "=" << eq + 1 << ";" << endl;
hessian_output << "v2";
hessianHelper(hessian_output, k, 1, output_type);
hessian_output << "=" << col_nb + 1 << ";" << endl;
hessian_output << "v2";
hessianHelper(hessian_output, k, 2, output_type);
hessian_output << "=";
d2->writeOutput(hessian_output, output_type, temporary_terms);
hessian_output << ";" << endl;
k++;
// Treating symetric elements
if (id1 != id2)
{
hessian_output << "v2";
hessianHelper(hessian_output, k, 0, output_type);
hessian_output << "=" << eq + 1 << ";" << endl;
hessian_output << "v2";
hessianHelper(hessian_output, k, 1, output_type);
hessian_output << "=" << col_nb_sym + 1 << ";" << endl;
hessian_output << "v2";
hessianHelper(hessian_output, k, 2, output_type);
hessian_output << "=v2";
hessianHelper(hessian_output, k-1, 2, output_type);
hessian_output << ";" << endl;
k++;
}
}
// Writing third derivatives
k = 0; // Keep the line of a 3rd derivative in v3
for (third_derivatives_type::const_iterator it = third_derivatives.begin();
it != third_derivatives.end(); it++)
{
int eq = it->first.first;
int var1 = it->first.second.first;
int var2 = it->first.second.second.first;
int var3 = it->first.second.second.second;
NodeID d3 = it->second;
int id1 = getDynJacobianCol(var1);
int id2 = getDynJacobianCol(var2);
int id3 = getDynJacobianCol(var3);
// Reference column number for the g3 matrix
int ref_col = id1 * hessianColsNbr + id2 * dynJacobianColsNbr + id3;
third_derivatives_output << " v3(" << ++k << ",:) = ["
<< eq + 1 << "," << ref_col + 1 << ",";
d3->writeOutput(third_derivatives_output, output_type, temporary_terms);
third_derivatives_output << "];" << endl;
// Compute the column numbers for the 5 other permutations of (id1,id2,id3) and store them in a set (to avoid duplicates if two indexes are equal)
set<int> cols;
cols.insert(id1 * hessianColsNbr + id3 * dynJacobianColsNbr + id2);
cols.insert(id2 * hessianColsNbr + id1 * dynJacobianColsNbr + id3);
cols.insert(id2 * hessianColsNbr + id3 * dynJacobianColsNbr + id1);
cols.insert(id3 * hessianColsNbr + id1 * dynJacobianColsNbr + id2);
cols.insert(id3 * hessianColsNbr + id2 * dynJacobianColsNbr + id1);
int k2 = 0; // Keeps the offset of the permutation relative to k
for (set<int>::iterator it2 = cols.begin(); it2 != cols.end(); it2++)
if (*it2 != ref_col)
third_derivatives_output << " v3(" << k + ++k2 << ",:) = ["
<< eq + 1 << "," << *it2 + 1 << ","
<< "v3(" << k << ",3)];" << endl;
k += k2;
}
if (!use_dll)
{
DynamicOutput << "%" << endl
<< "% Model equations" << endl
<< "%" << endl
<< endl
<< "residual = zeros(" << nrows << ", 1);" << endl
<< model_output.str() // Writing initialization instruction for matrix g1
<< "if nargout >= 2," << endl
<< " g1 = zeros(" << nrows << ", " << dynJacobianColsNbr << ");" << endl
<< endl
<< "%" << endl
<< "% Jacobian matrix" << endl
<< "%" << endl
<< endl
<< jacobian_output.str()
<< "end" << endl;
// Initialize g2 matrix
DynamicOutput << "if nargout >= 3," << endl
<< "%" << endl
<< "% Hessian matrix" << endl
<< "%" << endl
<< endl;
if (second_derivatives.size())
DynamicOutput << " v2 = zeros(" << NNZDerivatives[1] << ",3);" << endl
<< hessian_output.str()
<< " g2 = sparse(v2(:,1),v2(:,2),v2(:,3)," << nrows << "," << hessianColsNbr << ");" << endl;
else // Either hessian is all zero, or we didn't compute it
DynamicOutput << " g2 = sparse([],[],[]," << nrows << "," << hessianColsNbr << ");" << endl;
DynamicOutput << "end;" << endl;
// Initialize g3 matrix
DynamicOutput << "if nargout >= 4," << endl
<< "%" << endl
<< "% Third order derivatives" << endl
<< "%" << endl
<< endl;
int ncols = hessianColsNbr * dynJacobianColsNbr;
if (third_derivatives.size())
DynamicOutput << " v3 = zeros(" << NNZDerivatives[2] << ",3);" << endl
<< third_derivatives_output.str()
<< " g3 = sparse(v3(:,1),v3(:,2),v3(:,3)," << nrows << "," << ncols << ");" << endl;
else // Either 3rd derivatives is all zero, or we didn't compute it
DynamicOutput << " g3 = sparse([],[],[]," << nrows << "," << ncols << ");" << endl;
DynamicOutput << "end;" << endl;
}
else
{
DynamicOutput << "void Dynamic(double *y, double *x, int nb_row_x, double *params, int it_, double *residual, double *g1, double *v2)" << endl
<< "{" << endl
<< " double lhs, rhs;" << endl
<< endl
<< " /* Residual equations */" << endl
<< model_output.str()
<< " /* Jacobian */" << endl
<< " if (g1 == NULL)" << endl
<< " return;" << endl
<< " else" << endl
<< " {" << endl
<< jacobian_output.str()
<< " }" << endl;
if (second_derivatives.size())
{
DynamicOutput << " /* Hessian for endogenous and exogenous variables */" << endl
<< " if (v2 == NULL)" << endl
<< " return;" << endl
<< " else" << endl
<< " {" << endl
<< hessian_output.str()
<< " }" << endl;
}
DynamicOutput << "}" << endl << endl;
}
}
void
DynamicModel::writeOutput(ostream &output, const string &basename, bool block, bool byte_code, bool use_dll) const
{
/* Writing initialisation for M_.lead_lag_incidence matrix
M_.lead_lag_incidence is a matrix with as many columns as there are
endogenous variables and as many rows as there are periods in the
models (nbr of rows = M_.max_lag+M_.max_lead+1)
The matrix elements are equal to zero if a variable isn't present in the
model at a given period.
*/
output << "M_.lead_lag_incidence = [";
// Loop on endogenous variables
for (int endoID = 0; endoID < symbol_table.endo_nbr(); endoID++)
{
output << endl;
// Loop on periods
for (int lag = -max_endo_lag; lag <= max_endo_lead; lag++)
{
// Print variableID if exists with current period, otherwise print 0
try
{
int varID = getDerivID(symbol_table.getID(eEndogenous, endoID), lag);
output << " " << getDynJacobianCol(varID) + 1;
}
catch (UnknownDerivIDException &e)
{
output << " 0";
}
}
output << ";";
}
output << "]';" << endl;
// Write equation tags
output << "M_.equations_tags = {" << endl;
for (int i = 0; i < equation_tags.size(); i++)
output << " " << equation_tags[i].first + 1 << " , '"
<< equation_tags[i].second.first << "' , '"
<< equation_tags[i].second.second << "' ;" << endl;
output << "};" << endl;
//In case of sparse model, writes the block structure of the model
if (block)
{
//int prev_Simulation_Type=-1;
//bool skip_the_head;
int k=0;
int count_lead_lag_incidence = 0;
int max_lead, max_lag, max_lag_endo, max_lead_endo, max_lag_exo, max_lead_exo;
for (int j = 0;j < block_triangular.ModelBlock->Size;j++)
{
//For a block composed of a single equation determines wether we have to evaluate or to solve the equation
//skip_the_head=false;
k++;
count_lead_lag_incidence = 0;
int Block_size=block_triangular.ModelBlock->Block_List[j].Size;
max_lag =block_triangular.ModelBlock->Block_List[j].Max_Lag ;
max_lead=block_triangular.ModelBlock->Block_List[j].Max_Lead;
max_lag_endo =block_triangular.ModelBlock->Block_List[j].Max_Lag_Endo ;
max_lead_endo=block_triangular.ModelBlock->Block_List[j].Max_Lead_Endo;
max_lag_exo =block_triangular.ModelBlock->Block_List[j].Max_Lag_Exo ;
max_lead_exo=block_triangular.ModelBlock->Block_List[j].Max_Lead_Exo;
bool evaluate=false;
vector<int> exogenous;
vector<int>::iterator it_exogenous;
exogenous.clear();
ostringstream tmp_s, tmp_s_eq; tmp_s.str("");
tmp_s_eq.str("");
for (int i=0;i<block_triangular.ModelBlock->Block_List[j].Size;i++)
{
tmp_s << " " << block_triangular.ModelBlock->Block_List[j].Variable[i]+1;
tmp_s_eq << " " << block_triangular.ModelBlock->Block_List[j].Equation[i]+1;
}
for (int i=0;i<block_triangular.ModelBlock->Block_List[j].nb_exo;i++)
{
int ii=block_triangular.ModelBlock->Block_List[j].Exogenous[i];
for (it_exogenous=exogenous.begin();it_exogenous!=exogenous.end() && *it_exogenous!=ii;it_exogenous++) /*cout << "*it_exogenous=" << *it_exogenous << "\n"*/;
if (it_exogenous==exogenous.end() || exogenous.begin()==exogenous.end())
exogenous.push_back(ii);
}
output << "M_.block_structure.block(" << k << ").num = " << j+1 << ";\n";
output << "M_.block_structure.block(" << k << ").Simulation_Type = " << block_triangular.ModelBlock->Block_List[j].Simulation_Type << ";\n";
output << "M_.block_structure.block(" << k << ").maximum_lag = " << max_lag << ";\n";
output << "M_.block_structure.block(" << k << ").maximum_lead = " << max_lead << ";\n";
output << "M_.block_structure.block(" << k << ").maximum_endo_lag = " << max_lag_endo << ";\n";
output << "M_.block_structure.block(" << k << ").maximum_endo_lead = " << max_lead_endo << ";\n";
output << "M_.block_structure.block(" << k << ").maximum_exo_lag = " << max_lag_exo << ";\n";
output << "M_.block_structure.block(" << k << ").maximum_exo_lead = " << max_lead_exo << ";\n";
output << "M_.block_structure.block(" << k << ").endo_nbr = " << Block_size << ";\n";
output << "M_.block_structure.block(" << k << ").equation = [" << tmp_s_eq.str() << "];\n";
output << "M_.block_structure.block(" << k << ").variable = [" << tmp_s.str() << "];\n";
output << "M_.block_structure.block(" << k << ").exogenous = [";
int i=0;
for (it_exogenous=exogenous.begin();it_exogenous!=exogenous.end();it_exogenous++)
if (*it_exogenous>=0)
{
output << " " << *it_exogenous+1;
i++;
}
output << "];\n";
output << "M_.block_structure.block(" << k << ").exo_nbr = " << i << ";\n";
output << "M_.block_structure.block(" << k << ").exo_det_nbr = " << block_triangular.ModelBlock->Block_List[j].nb_exo_det << ";\n";
tmp_s.str("");
bool done_IM=false;
if (!evaluate)
{
output << "M_.block_structure.block(" << k << ").lead_lag_incidence = [];\n";
for (int l=-max_lag_endo;l<max_lead_endo+1;l++)
{
bool *tmp_IM;
tmp_IM=block_triangular.incidencematrix.Get_IM(l, eEndogenous);
if (tmp_IM)
{
for (int l_var=0;l_var<block_triangular.ModelBlock->Block_List[j].Size;l_var++)
{
for (int l_equ=0;l_equ<block_triangular.ModelBlock->Block_List[j].Size;l_equ++)
if (tmp_IM[block_triangular.ModelBlock->Block_List[j].Equation[l_equ]*symbol_table.endo_nbr()+block_triangular.ModelBlock->Block_List[j].Variable[l_var]])
{
count_lead_lag_incidence++;
if (tmp_s.str().length())
tmp_s << " ";
tmp_s << count_lead_lag_incidence;
done_IM=true;
break;
}
if (!done_IM)
tmp_s << " 0";
done_IM=false;
}
output << "M_.block_structure.block(" << k << ").lead_lag_incidence = [ M_.block_structure.block(" << k << ").lead_lag_incidence; " << tmp_s.str() << "];\n";
tmp_s.str("");
}
} }
else
{
bool done_some_where;
output << "M_.block_structure.block(" << k << ").lead_lag_incidence = [\n";
for (int l=-max_lag_endo;l<max_lead_endo+1;l++)
{
bool not_increm=true;
bool *tmp_IM;
tmp_IM=block_triangular.incidencematrix.Get_IM(l, eEndogenous);
int ii=j;
if (tmp_IM)
{
done_some_where = false;
while (ii-j<Block_size)
{
for (int l_var=0;l_var<block_triangular.ModelBlock->Block_List[ii].Size;l_var++)
{
for (int l_equ=0;l_equ<block_triangular.ModelBlock->Block_List[ii].Size;l_equ++)
if (tmp_IM[block_triangular.ModelBlock->Block_List[ii].Equation[l_equ]*symbol_table.endo_nbr()+block_triangular.ModelBlock->Block_List[ii].Variable[l_var]])
{
//if(not_increm && l==-max_lag)
count_lead_lag_incidence++;
not_increm=false;
if (tmp_s.str().length())
tmp_s << " ";
//tmp_s << count_lead_lag_incidence+(l+max_lag)*Block_size;
tmp_s << count_lead_lag_incidence;
done_IM=true;
break;
}
if (!done_IM)
tmp_s << " 0";
else
done_some_where = true;
done_IM=false;
}
ii++;
}
output << tmp_s.str() << "\n";
tmp_s.str("");
}
}
output << "];\n";
}
}
for (int j=-block_triangular.incidencematrix.Model_Max_Lag_Endo;j<=block_triangular.incidencematrix.Model_Max_Lead_Endo;j++)
{
bool* IM = block_triangular.incidencematrix.Get_IM(j, eEndogenous);
if (IM)
{
bool new_entry=true;
output << "M_.block_structure.incidence(" << block_triangular.incidencematrix.Model_Max_Lag_Endo+j+1 << ").lead_lag = " << j << ";\n";
output << "M_.block_structure.incidence(" << block_triangular.incidencematrix.Model_Max_Lag_Endo+j+1 << ").sparse_IM = [";
for (int i=0;i<symbol_table.endo_nbr()*symbol_table.endo_nbr();i++)
{
if (IM[i])
{
if (!new_entry)
output << " ; ";
else
output << " ";
output << i/symbol_table.endo_nbr()+1 << " " << i % symbol_table.endo_nbr()+1;
new_entry=false;
}
}
output << "];\n";
}
} }
// Writing initialization for some other variables
output << "M_.exo_names_orig_ord = [1:" << symbol_table.exo_nbr() << "];" << endl
<< "M_.maximum_lag = " << max_lag << ";" << endl
<< "M_.maximum_lead = " << max_lead << ";" << endl;
if (symbol_table.endo_nbr())
{
output << "M_.maximum_endo_lag = " << max_endo_lag << ";" << endl
<< "M_.maximum_endo_lead = " << max_endo_lead << ";" << endl
<< "oo_.steady_state = zeros(" << symbol_table.endo_nbr() << ", 1);" << endl;
}
if (symbol_table.exo_nbr())
{
output << "M_.maximum_exo_lag = " << max_exo_lag << ";" << endl
<< "M_.maximum_exo_lead = " << max_exo_lead << ";" << endl
<< "oo_.exo_steady_state = zeros(" << symbol_table.exo_nbr() << ", 1);" << endl;
}
if (symbol_table.exo_det_nbr())
{
output << "M_.maximum_exo_det_lag = " << max_exo_det_lag << ";" << endl
<< "M_.maximum_exo_det_lead = " << max_exo_det_lead << ";" << endl
<< "oo_.exo_det_steady_state = zeros(" << symbol_table.exo_det_nbr() << ", 1);" << endl;
}
if (symbol_table.param_nbr())
output << "M_.params = repmat(NaN," << symbol_table.param_nbr() << ", 1);" << endl;
// Write number of non-zero derivatives
output << "M_.NNZDerivatives = zeros(3, 1);" << endl
<< "M_.NNZDerivatives(1) = " << NNZDerivatives[0] << ";" << endl
<< "M_.NNZDerivatives(2) = " << NNZDerivatives[1] << ";" << endl
<< "M_.NNZDerivatives(3) = " << NNZDerivatives[2] << ";" << endl;
}
void
DynamicModel::evaluateJacobian(const eval_context_type &eval_context, jacob_map *j_m, bool dynamic)
{
int i=0;
int j=0;
bool *IM=NULL;
int a_variable_lag=-9999;
for (first_derivatives_type::iterator it = first_derivatives.begin();
it != first_derivatives.end(); it++)
{
//cout << "it->first.second=" << it->first.second << " variable_table.getSymbolID(it->first.second)=" << variable_table.getSymbolID(it->first.second) << " Type=" << variable_table.getType(it->first.second) << " eEndogenous=" << eEndogenous << " eExogenous=" << eExogenous << " variable_table.getLag(it->first.second)=" << variable_table.getLag(it->first.second) << "\n";
if (getTypeByDerivID(it->first.second) == eEndogenous)
{
NodeID Id = it->second;
double val = 0;
try
{
val = Id->eval(eval_context);
}
catch (ExprNode::EvalException &e)
{
cout << "evaluation of Jacobian failed for equation " << it->first.first+1 << " and variable " << symbol_table.getName(getSymbIDByDerivID(it->first.second)) << "(" << getLagByDerivID(it->first.second) << ") [" << getSymbIDByDerivID(it->first.second) << "] !" << endl;
Id->writeOutput(cout, oMatlabDynamicModelSparse, temporary_terms);
cout << "\n";
cerr << "DynamicModel::evaluateJacobian: evaluation of Jacobian failed for equation " << it->first.first+1 << " and variable " << symbol_table.getName(getSymbIDByDerivID(it->first.second)) << "(" << getLagByDerivID(it->first.second) << ")!" << endl;
}
int eq=it->first.first;
int var = symbol_table.getTypeSpecificID(getSymbIDByDerivID(it->first.second));///symbol_table.getID(eEndogenous,it->first.second);//variable_table.getSymbolID(it->first.second);
int k1 = getLagByDerivID(it->first.second);
if (a_variable_lag!=k1)
{
IM=block_triangular.incidencematrix.Get_IM(k1, eEndogenous);
a_variable_lag=k1;
}
if (k1==0 or !dynamic)
{
j++; (*j_m)[make_pair(eq,var)]+=val;
}
if (IM[eq*symbol_table.endo_nbr()+var] && (fabs(val) < cutoff))
{
if (block_triangular.bt_verbose)
cout << "the coefficient related to variable " << var << " with lag " << k1 << " in equation " << eq << " is equal to " << val << " and is set to 0 in the incidence matrix (size=" << symbol_table.endo_nbr() << ")\n";
block_triangular.incidencematrix.unfill_IM(eq, var, k1, eEndogenous);
i++;
}
}
}
//Get ride of the elements of the incidence matrix equal to Zero
IM=block_triangular.incidencematrix.Get_IM(0, eEndogenous);
for (int i=0;i<symbol_table.endo_nbr();i++)
for (int j=0;j<symbol_table.endo_nbr();j++)
if (IM[i*symbol_table.endo_nbr()+j])
if (first_derivatives.find(make_pair(i,getDerivID(symbol_table.getID(eEndogenous, j), 0)))==first_derivatives.end())
block_triangular.incidencematrix.unfill_IM(i, j, 0, eEndogenous);
if (i>0)
{
cout << i << " elements among " << first_derivatives.size() << " in the incidence matrices are below the cutoff (" << cutoff << ") and are discarded\n";
cout << "the contemporaneous incidence matrix has " << j << " elements\n";
}
}
void
DynamicModel::BlockLinear(Model_Block *ModelBlock)
{
int i,j,l,m,ll;
for (j = 0;j < ModelBlock->Size;j++)
{
if (ModelBlock->Block_List[j].Simulation_Type==SOLVE_BACKWARD_COMPLETE ||
ModelBlock->Block_List[j].Simulation_Type==SOLVE_FORWARD_COMPLETE)
{
ll=ModelBlock->Block_List[j].Max_Lag;
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[ll].size;i++)
{
int eq=ModelBlock->Block_List[j].IM_lead_lag[ll].Equ_Index[i];
int var=ModelBlock->Block_List[j].IM_lead_lag[ll].Var_Index[i];
//first_derivatives_type::const_iterator it=first_derivatives.find(make_pair(eq,variable_table.getID(var,0)));
first_derivatives_type::const_iterator it=first_derivatives.find(make_pair(eq,getDerivID(symbol_table.getID(eEndogenous, var),0)));
if (it!= first_derivatives.end())
{
NodeID Id = it->second;
set<pair<int, int> > endogenous;
Id->collectEndogenous(endogenous);
if (endogenous.size() > 0)
{
for (l=0;l<ModelBlock->Block_List[j].Size;l++)
{
if (endogenous.find(make_pair(ModelBlock->Block_List[j].Variable[l], 0)) != endogenous.end())
{
ModelBlock->Block_List[j].is_linear=false;
goto follow;
}
}
}
}
}
}
else if (ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_COMPLETE || ModelBlock->Block_List[j].Simulation_Type==SOLVE_TWO_BOUNDARIES_SIMPLE)
{
for (m=0;m<=ModelBlock->Block_List[j].Max_Lead+ModelBlock->Block_List[j].Max_Lag;m++)
{
int k1=m-ModelBlock->Block_List[j].Max_Lag;
for (i=0;i<ModelBlock->Block_List[j].IM_lead_lag[m].size;i++)
{
int eq=ModelBlock->Block_List[j].IM_lead_lag[m].Equ_Index[i];
int var=ModelBlock->Block_List[j].IM_lead_lag[m].Var_Index[i];
//first_derivatives_type::const_iterator it=first_derivatives.find(make_pair(eq,variable_table.getID(var,k1))); first_derivatives_type::const_iterator it=first_derivatives.find(make_pair(eq,getDerivID(symbol_table.getID(eEndogenous, var),k1)));
NodeID Id = it->second;
if (it!= first_derivatives.end())
{
set<pair<int, int> > endogenous;
Id->collectEndogenous(endogenous);
if (endogenous.size() > 0)
{
for (l=0;l<ModelBlock->Block_List[j].Size;l++)
{
if (endogenous.find(make_pair(ModelBlock->Block_List[j].Variable[l], k1)) != endogenous.end())
{
ModelBlock->Block_List[j].is_linear=false;
goto follow;
}
}
}
}
}
}
}
follow:
i=0;
}
}
map<pair<int, pair<int, int > >, NodeID>
DynamicModel::collect_first_order_derivatives_endogenous()
{
map<pair<int, pair<int, int > >, NodeID> endo_derivatives;
for (first_derivatives_type::iterator it2 = first_derivatives.begin();
it2 != first_derivatives.end(); it2++)
{
if (getTypeByDerivID(it2->first.second)==eEndogenous)
{
int eq = it2->first.first;
int var=symbol_table.getTypeSpecificID(getSymbIDByDerivID(it2->first.second));
int lag=getLagByDerivID(it2->first.second);
//if (lag==0)
endo_derivatives[make_pair(eq, make_pair(var, lag))] = it2->second;
}
}
return endo_derivatives;
}
void
DynamicModel::computingPass(bool jacobianExo, bool hessian, bool thirdDerivatives, bool paramsDerivatives,
const eval_context_type &eval_context, bool no_tmp_terms, bool block, bool use_dll)
{
assert(jacobianExo || !(hessian || thirdDerivatives || paramsDerivatives));
// Prepare for derivation
computeDerivIDs();
// Computes dynamic jacobian columns, must be done after computeDerivIDs()
computeDynJacobianCols(jacobianExo);
// Compute derivatives w.r. to all endogenous, and possibly exogenous and exogenous deterministic
set<int> vars;
for (deriv_id_table_t::const_iterator it = deriv_id_table.begin();
it != deriv_id_table.end(); it++)
{
SymbolType type = symbol_table.getType(it->first.first);
if (type == eEndogenous || (jacobianExo && (type == eExogenous || type == eExogenousDet)))
vars.insert(it->second);
}
// Launch computations
cout << "Computing dynamic model derivatives:" << endl
<< " - order 1" << endl;
computeJacobian(vars);
if (hessian)
{
cout << " - order 2" << endl;
computeHessian(vars);
}
if (paramsDerivatives)
{
cout << " - order 2 (derivatives of Jacobian w.r. to parameters)" << endl;
computeParamsDerivatives();
if (!no_tmp_terms)
computeParamsDerivativesTemporaryTerms();
}
if (thirdDerivatives)
{
cout << " - order 3" << endl;
computeThirdDerivatives(vars);
}
if (block)
{
BuildIncidenceMatrix();
jacob_map j_m;
evaluateJacobian(eval_context, &j_m, true);
if (block_triangular.bt_verbose)
{
cout << "The gross incidence matrix \n";
block_triangular.incidencematrix.Print_IM(eEndogenous);
}
t_etype equation_simulation_type;
map<pair<int, pair<int, int> >, NodeID> first_order_endo_derivatives = collect_first_order_derivatives_endogenous();
block_triangular.Normalize_and_BlockDecompose_Static_0_Model(j_m, equations, equation_simulation_type, first_order_endo_derivatives, mfs, cutoff);
BlockLinear(block_triangular.ModelBlock);
computeChainRuleJacobian(block_triangular.ModelBlock);
if (!no_tmp_terms)
computeTemporaryTermsOrdered(block_triangular.ModelBlock);
}
else
if (!no_tmp_terms)
computeTemporaryTerms(!use_dll);
}
void
DynamicModel::writeDynamicFile(const string &basename, bool block, bool bytecode, bool use_dll) const
{
int r;
if(block && bytecode)
{
writeModelEquationsCodeOrdered(basename + "_dynamic", block_triangular.ModelBlock, basename, map_idx);
block_triangular.Free_Block(block_triangular.ModelBlock);
block_triangular.incidencematrix.Free_IM();
//block_triangular.Free_IM_X(block_triangular.First_IM_X);
}
else if(block && !bytecode)
{#ifdef _WIN32
r = mkdir(basename.c_str());
#else
r = mkdir(basename.c_str(), 0777);
#endif
if (r < 0 && errno != EEXIST)
{
perror("ERROR");
exit(EXIT_FAILURE);
}
writeSparseDynamicMFile(basename + "_dynamic", basename);
block_triangular.Free_Block(block_triangular.ModelBlock);
block_triangular.incidencematrix.Free_IM();
//block_triangular.Free_IM_X(block_triangular.First_IM_X);
}
else if (use_dll)
writeDynamicCFile(basename + "_dynamic");
else
writeDynamicMFile(basename + "_dynamic");
}
void
DynamicModel::toStatic(StaticModel &static_model) const
{
// Convert model local variables (need to be done first)
for (map<int, NodeID>::const_iterator it = local_variables_table.begin();
it != local_variables_table.end(); it++)
static_model.AddLocalVariable(symbol_table.getName(it->first), it->second->toStatic(static_model));
// Convert equations
for (vector<BinaryOpNode *>::const_iterator it = equations.begin();
it != equations.end(); it++)
static_model.addEquation((*it)->toStatic(static_model));
// Convert auxiliary equations
for (deque<BinaryOpNode *>::const_iterator it = aux_equations.begin();
it != aux_equations.end(); it++)
static_model.addAuxEquation((*it)->toStatic(static_model));
}
void
DynamicModel::toStaticDll(StaticDllModel &static_model) const
{
// Convert model local variables (need to be done first)
for (map<int, NodeID>::const_iterator it = local_variables_table.begin();
it != local_variables_table.end(); it++)
static_model.AddLocalVariable(symbol_table.getName(it->first), it->second->toStatic(static_model));
// Convert equations
for (vector<BinaryOpNode *>::const_iterator it = equations.begin();
it != equations.end(); it++)
static_model.addEquation((*it)->toStatic(static_model));
}
void
DynamicModel::computeDerivIDs()
{
set<pair<int, int> > dynvars;
for(int i = 0; i < (int) equations.size(); i++)
equations[i]->collectVariables(eEndogenous, dynvars);
dynJacobianColsNbr = dynvars.size();
for(int i = 0; i < (int) equations.size(); i++)
{
equations[i]->collectVariables(eExogenous, dynvars);
equations[i]->collectVariables(eExogenousDet, dynvars);
equations[i]->collectVariables(eParameter, dynvars);
}
for(set<pair<int, int> >::const_iterator it = dynvars.begin();
it != dynvars.end(); it++)
{
int lag = it->second;
SymbolType type = symbol_table.getType(it->first);
// Setting maximum and minimum lags
if (max_lead < lag)
max_lead = lag;
else if (-max_lag > lag)
max_lag = -lag;
switch (type)
{
case eEndogenous:
if (max_endo_lead < lag)
max_endo_lead = lag;
else if (-max_endo_lag > lag)
max_endo_lag = -lag;
break;
case eExogenous:
if (max_exo_lead < lag)
max_exo_lead = lag;
else if (-max_exo_lag > lag)
max_exo_lag = -lag;
break;
case eExogenousDet:
if (max_exo_det_lead < lag)
max_exo_det_lead = lag;
else if (-max_exo_det_lag > lag)
max_exo_det_lag = -lag;
break;
default:
break;
}
// Create a new deriv_id
int deriv_id = deriv_id_table.size();
deriv_id_table[*it] = deriv_id;
inv_deriv_id_table.push_back(*it);
}
}
SymbolType
DynamicModel::getTypeByDerivID(int deriv_id) const throw (UnknownDerivIDException)
{
return symbol_table.getType(getSymbIDByDerivID(deriv_id));
}
int
DynamicModel::getLagByDerivID(int deriv_id) const throw (UnknownDerivIDException)
{
if (deriv_id < 0 || deriv_id >= (int) inv_deriv_id_table.size())
throw UnknownDerivIDException();
return inv_deriv_id_table[deriv_id].second;
}
int
DynamicModel::getSymbIDByDerivID(int deriv_id) const throw (UnknownDerivIDException)
{
if (deriv_id < 0 || deriv_id >= (int) inv_deriv_id_table.size())
throw UnknownDerivIDException();
return inv_deriv_id_table[deriv_id].first;
}
intDynamicModel::getDerivID(int symb_id, int lag) const throw (UnknownDerivIDException)
{
deriv_id_table_t::const_iterator it = deriv_id_table.find(make_pair(symb_id, lag));
if (it == deriv_id_table.end())
throw UnknownDerivIDException();
else
return it->second;
}
void
DynamicModel::computeDynJacobianCols(bool jacobianExo)
{
/* Sort the dynamic endogenous variables by lexicographic order over (lag, type_specific_symbol_id)
and fill the dynamic columns for exogenous and exogenous deterministic */
map<pair<int, int>, int> ordered_dyn_endo;
for (deriv_id_table_t::const_iterator it = deriv_id_table.begin();
it != deriv_id_table.end(); it++)
{
const int &symb_id = it->first.first;
const int &lag = it->first.second;
const int &deriv_id = it->second;
SymbolType type = symbol_table.getType(symb_id);
int tsid = symbol_table.getTypeSpecificID(symb_id);
switch (type)
{
case eEndogenous:
ordered_dyn_endo[make_pair(lag, tsid)] = deriv_id;
break;
case eExogenous:
// At this point, dynJacobianColsNbr contains the number of dynamic endogenous
if (jacobianExo)
dyn_jacobian_cols_table[deriv_id] = dynJacobianColsNbr + tsid;
break;
case eExogenousDet:
// At this point, dynJacobianColsNbr contains the number of dynamic endogenous
if (jacobianExo)
dyn_jacobian_cols_table[deriv_id] = dynJacobianColsNbr + symbol_table.exo_nbr() + tsid;
break;
case eParameter:
// We don't assign a dynamic jacobian column to parameters
break;
default:
// Shut up GCC
cerr << "DynamicModel::computeDynJacobianCols: impossible case" << endl;
exit(EXIT_FAILURE);
}
}
// Fill in dynamic jacobian columns for endogenous
int sorted_id = 0;
for (map<pair<int, int>, int>::const_iterator it = ordered_dyn_endo.begin();
it != ordered_dyn_endo.end(); it++)
dyn_jacobian_cols_table[it->second] = sorted_id++;
// Set final value for dynJacobianColsNbr
if (jacobianExo)
dynJacobianColsNbr += symbol_table.exo_nbr() + symbol_table.exo_det_nbr();
}
int
DynamicModel::getDynJacobianCol(int deriv_id) const throw (UnknownDerivIDException)
{
map<int, int>::const_iterator it = dyn_jacobian_cols_table.find(deriv_id);
if (it == dyn_jacobian_cols_table.end())
throw UnknownDerivIDException();
else
return it->second;
}
void
DynamicModel::computeChainRuleJacobian(Model_Block *ModelBlock)
{
map<int, NodeID> recursive_variables;
first_chain_rule_derivatives.clear();
for(int blck = 0; blck<ModelBlock->Size; blck++)
{
recursive_variables.clear();
if (ModelBlock->Block_List[blck].Simulation_Type==SOLVE_TWO_BOUNDARIES_COMPLETE or ModelBlock->Block_List[blck].Simulation_Type==SOLVE_TWO_BOUNDARIES_SIMPLE)
{
ModelBlock->Block_List[blck].Chain_Rule_Derivatives->clear();
for(int i = 0; i < ModelBlock->Block_List[blck].Nb_Recursives; i++)
{
if (ModelBlock->Block_List[blck].Equation_Type[i] == E_EVALUATE_S)
recursive_variables[getDerivID(symbol_table.getID(eEndogenous, ModelBlock->Block_List[blck].Variable[i]), 0)] = ModelBlock->Block_List[blck].Equation_Normalized[i];
else
recursive_variables[getDerivID(symbol_table.getID(eEndogenous, ModelBlock->Block_List[blck].Variable[i]), 0)] = equations[ModelBlock->Block_List[blck].Equation[i]];
}
map<pair<pair<int, pair<int, int> >, pair<int, int> >, int> Derivatives = block_triangular.get_Derivatives(ModelBlock, blck);
map<pair<pair<int, pair<int, int> >, pair<int, int> >, int>::const_iterator it = Derivatives.begin();
//#pragma omp parallel for shared(it, blck)
for(int i=0; i<(int)Derivatives.size(); i++)
{
int Deriv_type = it->second;
pair<pair<int, pair<int, int> >, pair<int, int> > it_l(it->first);
it++;
int lag = it_l.first.first;
int eq = it_l.first.second.first;
int var = it_l.first.second.second;
int eqr = it_l.second.first;
int varr = it_l.second.second;
if(Deriv_type == 0)
{
first_chain_rule_derivatives[make_pair(eqr, make_pair(varr, lag))] = first_derivatives[make_pair(eqr, getDerivID(symbol_table.getID(eEndogenous, varr), lag))];
}
else if (Deriv_type == 1)
{
first_chain_rule_derivatives[make_pair(eqr, make_pair(varr, lag))] = ModelBlock->Block_List[blck].Equation_Normalized[eq]->getChainRuleDerivative(getDerivID(symbol_table.getID(eEndogenous, varr), lag), recursive_variables);
}
else if (Deriv_type == 2)
{
if(ModelBlock->Block_List[blck].Equation_Type[eq] == E_EVALUATE_S && eq<ModelBlock->Block_List[blck].Nb_Recursives)
first_chain_rule_derivatives[make_pair(eqr, make_pair(varr, lag))] = ModelBlock->Block_List[blck].Equation_Normalized[eq]->getChainRuleDerivative(getDerivID(symbol_table.getID(eEndogenous, varr), lag), recursive_variables);
else
first_chain_rule_derivatives[make_pair(eqr, make_pair(varr, lag))] = equations[eqr]->getChainRuleDerivative(getDerivID(symbol_table.getID(eEndogenous, varr), lag), recursive_variables);
}
ModelBlock->Block_List[blck].Chain_Rule_Derivatives->push_back(make_pair( make_pair(lag, make_pair(eq, var)), make_pair(eqr, varr)));
}
}
else if( ModelBlock->Block_List[blck].Simulation_Type==SOLVE_BACKWARD_SIMPLE or ModelBlock->Block_List[blck].Simulation_Type==SOLVE_FORWARD_SIMPLE
or ModelBlock->Block_List[blck].Simulation_Type==SOLVE_BACKWARD_COMPLETE or ModelBlock->Block_List[blck].Simulation_Type==SOLVE_FORWARD_COMPLETE)
{
ModelBlock->Block_List[blck].Chain_Rule_Derivatives->clear();
for(int i = 0; i < ModelBlock->Block_List[blck].Nb_Recursives; i++)
{
if (ModelBlock->Block_List[blck].Equation_Type[i] == E_EVALUATE_S)
recursive_variables[getDerivID(symbol_table.getID(eEndogenous, ModelBlock->Block_List[blck].Variable[i]), 0)] = ModelBlock->Block_List[blck].Equation_Normalized[i];
else
recursive_variables[getDerivID(symbol_table.getID(eEndogenous, ModelBlock->Block_List[blck].Variable[i]), 0)] = equations[ModelBlock->Block_List[blck].Equation[i]];
}
for(int eq = ModelBlock->Block_List[blck].Nb_Recursives; eq < ModelBlock->Block_List[blck].Size; eq++)
{
int eqr = ModelBlock->Block_List[blck].Equation[eq];
for(int var = ModelBlock->Block_List[blck].Nb_Recursives; var < ModelBlock->Block_List[blck].Size; var++)
{
int varr = ModelBlock->Block_List[blck].Variable[var];
NodeID d1 = equations[eqr]->getChainRuleDerivative(getDerivID(symbol_table.getID(eEndogenous, varr), 0), recursive_variables); if (d1 == Zero)
continue;
first_chain_rule_derivatives[make_pair(eqr, make_pair(varr, 0))] = d1;
ModelBlock->Block_List[blck].Chain_Rule_Derivatives->push_back(make_pair( make_pair(0, make_pair(eq, var)), make_pair(eqr, varr)));
}
}
}
}
}
void
DynamicModel::computeParamsDerivatives()
{
for (deriv_id_table_t::const_iterator it = deriv_id_table.begin();
it != deriv_id_table.end(); it++)
{
if (symbol_table.getType(it->first.first) != eParameter)
continue;
int param = it->second;
for (int eq = 0; eq < (int) equations.size(); eq++)
{
NodeID d1 = equations[eq]->getDerivative(param);
if (d1 == Zero)
continue;
residuals_params_derivatives[make_pair(eq, param)] = d1;
}
for (first_derivatives_type::const_iterator it2 = first_derivatives.begin();
it2 != first_derivatives.end(); it2++)
{
int eq = it2->first.first;
int var = it2->first.second;
NodeID d1 = it2->second;
NodeID d2 = d1->getDerivative(param);
if (d2 == Zero)
continue;
jacobian_params_derivatives[make_pair(eq, make_pair(var, param))] = d2;
}
}
}
void
DynamicModel::computeParamsDerivativesTemporaryTerms()
{
map<NodeID, int> reference_count;
params_derivs_temporary_terms.clear();
for (first_derivatives_type::iterator it = residuals_params_derivatives.begin();
it != residuals_params_derivatives.end(); it++)
it->second->computeTemporaryTerms(reference_count, params_derivs_temporary_terms, true);
for (second_derivatives_type::iterator it = jacobian_params_derivatives.begin();
it != jacobian_params_derivatives.end(); it++)
it->second->computeTemporaryTerms(reference_count, params_derivs_temporary_terms, true);
}
void
DynamicModel::writeParamsDerivativesFile(const string &basename) const
{
if (!residuals_params_derivatives.size()
&& !jacobian_params_derivatives.size())
return;
string filename = basename + "_params_derivs.m";
ofstream paramsDerivsFile;
paramsDerivsFile.open(filename.c_str(), ios::out | ios::binary);
if (!paramsDerivsFile.is_open())
{
cerr << "ERROR: Can't open file " << filename << " for writing" << endl;
exit(EXIT_FAILURE);
}
paramsDerivsFile << "function [rp, gp] = " << basename << "_params_derivs(y, x, params, it_)" << endl
<< "%" << endl
<< "% Warning : this file is generated automatically by Dynare" << endl
<< "% from model file (.mod)" << endl << endl;
writeTemporaryTerms(params_derivs_temporary_terms, paramsDerivsFile, oMatlabDynamicModel);
// Write parameter derivative
paramsDerivsFile << "rp = zeros(" << equation_number() << ", "
<< symbol_table.param_nbr() << ");" << endl;
for (first_derivatives_type::const_iterator it = residuals_params_derivatives.begin();
it != residuals_params_derivatives.end(); it++)
{
int eq = it->first.first;
int param = it->first.second;
NodeID d1 = it->second;
int param_col = symbol_table.getTypeSpecificID(getSymbIDByDerivID(param)) + 1;
paramsDerivsFile << "rp(" << eq+1 << ", " << param_col << ") = ";
d1->writeOutput(paramsDerivsFile, oMatlabDynamicModel, params_derivs_temporary_terms);
paramsDerivsFile << ";" << endl;
}
// Write jacobian derivatives
paramsDerivsFile << "gp = zeros(" << equation_number() << ", " << dynJacobianColsNbr << ", "
<< symbol_table.param_nbr() << ");" << endl;
for (second_derivatives_type::const_iterator it = jacobian_params_derivatives.begin();
it != jacobian_params_derivatives.end(); it++)
{
int eq = it->first.first;
int var = it->first.second.first;
int param = it->first.second.second;
NodeID d2 = it->second;
int var_col = getDynJacobianCol(var) + 1;
int param_col = symbol_table.getTypeSpecificID(getSymbIDByDerivID(param)) + 1;
paramsDerivsFile << "gp(" << eq+1 << ", " << var_col << ", " << param_col << ") = ";
d2->writeOutput(paramsDerivsFile, oMatlabDynamicModel, params_derivs_temporary_terms);
paramsDerivsFile << ";" << endl;
}
paramsDerivsFile.close();
}
void
DynamicModel::writeChainRuleDerivative(ostream &output, int eqr, int varr, int lag,
ExprNodeOutputType output_type,
const temporary_terms_type &temporary_terms) const
{
map<pair<int, pair<int, int> >, NodeID>::const_iterator it = first_chain_rule_derivatives.find(make_pair(eqr, make_pair(varr, lag)));
if (it != first_chain_rule_derivatives.end())
(it->second)->writeOutput(output, output_type, temporary_terms);
else
output << 0;
}
void
DynamicModel::writeLatexFile(const string &basename) const
{
writeLatexModelFile(basename + "_dynamic.tex", oLatexDynamicModel);
}
void
DynamicModel::jacobianHelper(ostream &output, int eq_nb, int col_nb, ExprNodeOutputType output_type) const
{
output << LEFT_ARRAY_SUBSCRIPT(output_type);
if (IS_MATLAB(output_type))
output << eq_nb + 1 << ", " << col_nb + 1;
else
output << eq_nb + col_nb * equations.size();
output << RIGHT_ARRAY_SUBSCRIPT(output_type);
}
void
DynamicModel::hessianHelper(ostream &output, int row_nb, int col_nb, ExprNodeOutputType output_type) const
{
output << LEFT_ARRAY_SUBSCRIPT(output_type);
if (IS_MATLAB(output_type))
output << row_nb + 1 << ", " << col_nb + 1;
else
output << row_nb + col_nb * NNZDerivatives[1];
output << RIGHT_ARRAY_SUBSCRIPT(output_type);
}
void
DynamicModel::substituteLeadGreaterThanTwo()
{
ExprNode::subst_table_t subst_table;
vector<BinaryOpNode *> neweqs;
// Substitute in model local variables
for(map<int, NodeID>::iterator it = local_variables_table.begin();
it != local_variables_table.end(); it++)
it->second = it->second->substituteLeadGreaterThanTwo(subst_table, neweqs);
// Substitute in equations
for(int i = 0; i < (int) equations.size(); i++)
{
BinaryOpNode *substeq = dynamic_cast<BinaryOpNode *>(equations[i]->substituteLeadGreaterThanTwo(subst_table, neweqs));
assert(substeq != NULL);
equations[i] = substeq;
}
// Add new equations
for(int i = 0; i < (int) neweqs.size(); i++)
addEquation(neweqs[i]);
// Add the new set of equations at the *beginning* of aux_equations
copy(neweqs.rbegin(), neweqs.rend(), front_inserter(aux_equations));
if (neweqs.size() > 0)
cout << "Substitution of leads >= 2: added " << neweqs.size() << " auxiliary variables and equations." << endl;
}
void
DynamicModel::substituteLagGreaterThanTwo()
{
ExprNode::subst_table_t subst_table;
vector<BinaryOpNode *> neweqs;
// Substitute in model local variables
for(map<int, NodeID>::iterator it = local_variables_table.begin();
it != local_variables_table.end(); it++)
it->second = it->second->substituteLagGreaterThanTwo(subst_table, neweqs);
// Substitute in equations
for(int i = 0; i < (int) equations.size(); i++)
{
BinaryOpNode *substeq = dynamic_cast<BinaryOpNode *>(equations[i]->substituteLagGreaterThanTwo(subst_table, neweqs));
assert(substeq != NULL);
equations[i] = substeq;
}
// Add new equations
for(int i = 0; i < (int) neweqs.size(); i++)
addEquation(neweqs[i]);
// Add the new set of equations at the *beginning* of aux_equations
copy(neweqs.rbegin(), neweqs.rend(), front_inserter(aux_equations));
if (neweqs.size() > 0)
cout << "Substitution of lags >= 2: added " << neweqs.size() << " auxiliary variables and equations." << endl;
}
void
DynamicModel::fillEvalContext(eval_context_type &eval_context) const
{
// First, auxiliary variables
for(deque<BinaryOpNode *>::const_iterator it = aux_equations.begin();
it != aux_equations.end(); it++)
{
assert((*it)->get_op_code() == oEqual);
VariableNode *auxvar = dynamic_cast<VariableNode *>((*it)->get_arg1());
assert(auxvar != NULL);
try
{
double val = (*it)->get_arg2()->eval(eval_context);
eval_context[auxvar->get_symb_id()] = val;
}
catch(ExprNode::EvalException &e)
{
// Do nothing
}
}
// Second, model local variables
for(map<int, NodeID>::const_iterator it = local_variables_table.begin();
it != local_variables_table.end(); it++)
{
try
{
const NodeID expression = it->second;
double val = expression->eval(eval_context);
eval_context[it->first] = val;
}
catch(ExprNode::EvalException &e)
{
// Do nothing
}
}
}