Temporary terms need to be computed per equation (as was done previously), and
not simply per block.
It’s necessary to track temporary terms per equation, because some equations
are evaluated instead of solved, and an equation E1 may depend on the value of
an endogenous Y computed by a previously evaluated equation E2; in this case,
if some temporary term TT of equation E2 contains Y, then TT needs to be
computed after E1, but before E2.
In particular, in dynamic models, temporary terms are now computed for
derivatives w.r.t. exogenous, and also w.r.t. endogenous variables that do not
belong to the block.
Also remove a message about elements below the cutoff that was no longer
correct (elements below the cutoff have no impact on the incidence matrix
outside of normalization).
By the way, remove the BlockType stuff which was purely informative (and it’s
not worth carrying over prologue and epilogue information just for that).
Previously, the cutoff option would also impact the block decomposition itself,
since it would had an influence on the incidence matrix used for computing the
blocks and their derivatives.
The problem is that, in the general case, it’s quite possible that an element
of the numerical Jacobian be zero at the evaluation point, while being quite
different from zero along the simulation path. A typical example is an
expression of the type x*y, where y is an endogenous and x is an exogenous not
present in the initval block (and hence initialized to zero).
It has been superseded by ModelTree::blocks_derivatives.
By the way, fix the initial number of non-zero elements in sparse Jacobian.
Also avoid computing suboptimal temporary terms.
— use a std::map for storing block derivatives
— remove redundant ModelTree::first_chain_rule_derivatives structure
— remove unused codepaths in StaticModel
— DynamicModel: simplify code that determines the type of derivatives in a
block. We now use a slightly different categorization.
— by the way, fix the max lead/lag information for blocks that are obtained via
merging. A workaround was previously implemented in
DynamicModel::get_Derivative(), but it is no longer needed with this fix.
Since bef537d40a, constant equations were not
simplified as soon as they had a tag attached.
But this is too wide a restriction. In particular, this breaks the trend
component models which have a target that is set to a constant.
So we now only skip the replacement in the case where there is an “mcp” tag.
Ref. dynare#1697
This was only adding unneeded complexity, for no clear reason (we’re very far
from reaching 2³¹ equations, and if we wanted to support models that large, it
would be better to use long integers to avoid being limited to 2³²).
— return output arguments on the left-hand side
— do not pass class members as input/output arguments
By the way, fix a (benign) vector allocation bug in
{Static,Dynamic}Model::computeChainRuleJacobian().
Since commit 9c9e8f816f, it’s the information
from the original model which was in this field, which is not what is expected.
By the way, do not output this field (and the related M_.hessian_eq_zero) when
the Hessian is not computed by the preprocessor (i.e. in practice for perfect
foresight), since they would otherwise contain incorrect information.
Ref. dynare#1681
Auxiliary equations appearing in set_auxiliary_variables.m and
dynamic_set_auxiliary_series.m need to appear in recursive ordering, since
those files are used for sequential evaluation.
Previously, the recursive ordering was guaranteed by a set of ad hoc rules and
workarounds, but that would not cover certain edge cases.
With this commit, the recursive ordering is systematically computed, using a
topological sort on the directed acyclic graph whose vertices are auxiliary
equations and whose edges are dependency relationships.
Closes: #22
Allows for the inclusion/exclusion of a set of equations, specified either on the command line or in a text file.
If the equation has a single endogenous variable on the LHS, then the equation is moved. If not, if the equation has an `endogenous` tag then that variable is removed along with this equation. If not, then an error is thrown.
As a command line argument, `exclude_eqs` can take the form (same syntax for `include_eqs`):
* `exclude_eqs=eq1 to remove all equations declared as `[name=eq1]`
* `exclude_eqs=[eq 1, eq 2]` to remove all equations declared as `[name=eq 1]` or `[name=eq 2]`
* `exclude_eqs=[tagname=X]` to remove all equations declared as `[tagname=X]`
* `exclude_eqs=[tagname=(X, Y)]` to remove all equations declared as `[tagname=X]` or `[tagname=Y]`
When declared in a file, the file should be of the form:
```
eq 1
eq 2
```
to remove all equations declared as `[name=eq 1]` or `[name=eq 2]`.
It should be of the form:
```
tagname=
X
Y
```
to remove all equations declared as `[tagname=X]` or `[tagname=Y]`.
In many cases, they can be replaced by the curly braces syntax.
Otherwise, we can now use the pair() and tuple() constructors, without the need
to specify template parameters, thanks to class template argument
deduction (new in C++17).
Also, no longer compute two times symmetric elements in derivation w.r.t.
parameters at order 2, for consistency with derivation w.r.t. endogenous.
It is therefore now necessary to duplicate them in the output to keep behavior
unchanged.
New options "mexext" and "matlabroot" are introduced, so that the preprocessor
knows where to find MATLAB and which architecture to compile for.
Only recent gcc is now supported. A set of optimization flags is used so that
compilation goes reasonably fast on large models.
Consequently, options "msvc", "mingw" and "cygwin" have been removed.
In particular, it is necessary to turn back DataTree::AddVariable() into a
non-virtual method, because it is called from DataTree's constructor. Enforcing
the absence of leads/lags is now done using a new boolean DataTree::is_static.
Take advantage of the new copy constructor for handling
PlannerObjectiveStatement more elegantly.
Unfortunately it is not possible to implement *move* constructor / assigment
operators, because the reference ExprNode::datatree is not mutable.
aux_equations only contain the definition of auxiliary variables, and
may diverge from those in the main model (equations), if other model
transformations applied subsequently. This is not a problem, since
aux_equations is only used for regenerating the values of auxiliaries
given the others.
For example, such a divergence appears when there is an expectation
operator in a ramsey model, see
tests/optimal_policy/nk_ramsey_expectation.mod */
This table serves no useful purpose. It is better to append auxiliary equations
at the time they are created, to avoid messing with the recursive ordering.
In the presence of MLVs, the temporary terms indexing was corrupted. The code
was using the implicit assumption that the ExprNodeLess ordering was giving the
same ordering as the temporary terms indexes ordering. But MLVs can be higher
in ExprNodeLess ordering than some other temporary terms, while they have the
lowest temporary terms index, hence the bug.
Fix this by no longer relying on the ExprNodeLess ordering, and rather use a
full map<ExprNode *, int> for ModelTree::temporary_terms_idxs. By the way,
simplify the code by removing a few useless data structures (e.g.
ModelTree::temporary_terms_idxs_*).
Sébastien Villemot
Houtan Bastani
Ferhat Mihoubi