This is a Fortran 2008 rewrite of disclyap_fast.m.

Closes: #1737

Sébastien Villemot authored Jun 28, 2019 and Stéphane Adjemian committed Dec 20, 2019

It applies the approximated policy function to a set of particles, using
Dynare++ routines.

There is support for parallelization, using Dynare++ multithreading
model (itself based on C++11 threads; we don’t use OpenMP because it is
incompatible with MKL). For the time being, default to a single thread. This
should be later refined through empirical testing.

In particular, if either MATLAB or Octave is missing, one needs to pass either
--disable-matlab or --disable-octave.

Moreover, several new configure flags have been introduced for disabling some
components:
--disable-doc
--disable-dynare++
--disable-mex-dynare++
--disable-mex-ms-sbvar
--disable-mex-kalman-steady-state

It constructs the stacked residuals and jacobian of the perfect foresight
problem.

It is an almost perfect replacement for the perfect_foresight_problem.m
routine, while being much more efficient.

Note however that the DLL never return complex numbers (it instead puts NaNs at
the place where there would have been complex). This may create problems for
some MOD files; the algorithms will need to be adapted to use a more
line-search method.

It's actually not needed. The QZ complex decomposition can be obtained in
Octave on real matrices by casting them to the complex type.

These functions are all included in Octave 4.2.

The function is now an Octave builtin.

- no longer use OCTAVE_QUIT in MEX files; this seems now only possible in
  oct-files. (Ref #304)
- do not build linsolve.oct on Octave >= 3.8, it is available natively.
- do not add strjoin.m to the patch on Octave >= 3.8, it is available natively.
- default_save_options has been renamed save_default_options.

Does not seem very robust (fails on fs2000), need to investigate why.

Closes: #273

equations are approximated at order two around the deterministic steady state.

Added new mex file for computing Quasi Monte-Carlo sequences (Sobol), texinfo headers and unitary tests are available in <DYNARE_PATH>/matlab/qmc_sequence.m.

- remove the dirty vpath hack
- factorize between the two estimation DLLs
- factorize within SWZ makefile

The mex works with octave but not with matlab (lapack issue again ;-).

Added the Slicot library (distributed under the GNU General Public Licence at  www.slicot.org). This library provides routines
for the design and analysis of control systems. Adapted the building system to handle fortran 77 source files.

New oct-file "qzcomplex" for bringing the complex QZ decomposition to Octave. Fixes issues with partial information under Octave.

* added dynare_simul_ DLL
* factorized in a library object the Dynare++ code shared between gensylv, k_order_perturbation and dynare_simul_


git-svn-id: https://www.dynare.org/svn/dynare/trunk@3114 ac1d8469-bf42-47a9-8791-bf33cf982152

git-svn-id: https://www.dynare.org/svn/dynare/trunk@3070 ac1d8469-bf42-47a9-8791-bf33cf982152

Build system: refactored the build system for MEX files, using two sub-packages with their own configure scripts


git-svn-id: https://www.dynare.org/svn/dynare/trunk@3019 ac1d8469-bf42-47a9-8791-bf33cf982152