Michel Juillard
41e027aee3
added steady option stop_on_error. stop_on_error = 0 permits to
...
continue (for example, when chaining homotopy steps), even if the
steady state computation was not successful.
2012-04-16 22:40:50 +02:00
Houtan Bastani
08bcf39fe7
estimation: clear subsamples substructure whenever a subsamples statement is encountered
2012-04-03 18:33:23 +02:00
Houtan Bastani
7e6bd80f17
estimation: remove s from parameters substructure
2012-04-03 18:33:23 +02:00
Houtan Bastani
31e0bc2e1e
estimation: simplify global_initialization
2012-04-03 18:33:23 +02:00
Houtan Bastani
9741b3b3ec
estimation: reorganize options output
2012-03-30 15:37:15 +02:00
Houtan Bastani
93db8937ab
estimation: reorganize Matlab structure output
2012-03-30 15:37:14 +02:00
Stéphane Adjemian (Charybdis)
6c6b62937f
Added smooth resampling.
2012-03-30 11:47:24 +02:00
Houtan Bastani
7feb27fe1c
estimation: rewrite options output
2012-03-29 16:45:46 +02:00
Houtan Bastani
d54e4c4b17
estimation: rewrite prior output
2012-03-29 16:07:01 +02:00
Houtan Bastani
46a655fbd5
estimation: rewrite subsamples statement as an actual statement
2012-03-28 18:51:52 +02:00
Michel Juillard
a899594f3d
initializing globals ys0_ ex0_ ex_det0_
2012-03-17 20:34:58 +01:00
Michel Juillard
d6f8e252e5
Merge remote-tracking branch 'origin/master' into dr1break
2012-03-15 16:18:37 +01:00
Stéphane Adjemian (Charybdis)
c5fa0b82a9
Merge remote-tracking branch 'marco/master'
2012-03-14 12:49:56 +01:00
Michel Juillard
1fcf708b80
Merge remote-tracking branch 'origin/master' into dr1break
2012-03-10 18:21:14 +01:00
Stéphane Adjemian (Charybdis)
323e6fdf4f
Added a second gstep option used as a parameter for the routine computing the hessian matrix.
...
In some cases, for instance for the non linear filters, it helps to reduce this new gstep parameter
to get a positive definite hessian matrix. options_.gstep is now a 2*1 vector. The first element is
the old gstep parameter, the second element is the new gstep parameter. The step defined for the
computation of the hessian matrix is now:
h1=max(abs(x),sqrt(gstep(1))*ones(n,1))*eps^(1/6)*gstep(2);
2012-03-09 16:36:26 +01:00
Stéphane Adjemian (Charybdis)
ec9b47426a
Added an option for the threshold level of neff in the generic case of the resampling routines.
2012-03-09 12:44:25 +01:00
Houtan Bastani
ec2528ae9b
estimation: fixes to options statement
2012-03-09 11:47:25 +01:00
Houtan Bastani
dd67a81e57
estimation: place parameter priors in their own substructure
2012-03-09 11:46:09 +01:00
Stéphane Adjemian (Charybdis)
464dcb651e
Changed defaults related to extended path. Removed some unused options.
2012-03-08 12:45:53 +01:00
Ferhat Mihoubi
8ac14f8c80
Adds fixed point solvers for Sylvester and Lyapunov equations
2012-03-06 12:03:23 +01:00
Stéphane Adjemian (Charybdis)
a2f3a53634
Parallelization of local_state_space_iteration_2 (used in non linear filters).
2012-03-05 23:11:49 +01:00
Stéphane Adjemian (Charybdis)
cf19df92ef
Added an option fot the mode_check plots for defining the size of the neighbourhood around the estimated posterior mode.
2012-03-05 15:03:55 +01:00
Stéphane Adjemian (Charybdis)
83b536c618
Moved CMAES options in global_initialization, so that the options of this routine can be modified by
...
the user by writting in options_.cmaes.
2012-03-04 22:13:00 +01:00
Stéphane Adjemian (Charybdis)
724b99412f
Added option for kitagawa/stratified approach in traditional resampling.
2012-03-04 22:12:59 +01:00
Stéphane Adjemian (Charybdis)
fb4def04bd
Fixed bug related to the name of the sub-structure for the non linear filters.
2012-03-04 22:12:59 +01:00
Marco Ratto
c848b3337a
added nodisplay and graph_format options and applied to prior plots and graphs in estimstion routine. Tex tables also splitted when more than 50 params are estimated.
...
nograph: no figure is done
nodisplay: plots are done behind the scenes
graph_format: default only eps figs are saved. it accepts cell vector like {'eps','pdf','fig'}
2012-02-13 11:58:28 +01:00
Stéphane Adjemian (Charybdis)
87f4dad51a
Changed default value of use_bytecode option. By default, the bytecode solver is not used.
...
Note that, even for small models, there is a (very) substantial gain in using the use_dll option.
2012-02-04 16:56:09 +01:00
Stéphane Adjemian (Charybdis)
704b0c9659
Added an option specifying if the bytecode solver has to be used first.
2012-02-04 16:26:22 +01:00
Stéphane Adjemian (Charybdis)
8a35ee7363
Added an option to skip the test on the number of periods over which the perfect foresight models are solved.
2012-02-04 16:19:15 +01:00
Stéphane Adjemian (Charybdis)
64ebd1d0d7
Changed the default value of options_.ep.fp. Test the stability of the solution only for the first period (when the value of periods is increased).
2012-02-03 14:05:06 +01:00
Stéphane Adjemian (Charybdis)
f63ce01859
Put debug and memory modes in options_.ep.
2012-01-21 14:13:31 +01:00
Stéphane Adjemian (Charybdis)
58f4feb6ad
Cosmetic change + Added scramble mode (possibility to add noise in the future).
2012-01-20 16:40:17 +01:00
Michel Juillard
f9d2dec97f
preparing dsge_loglikelihood.m for future integration of estimation DLL
2012-01-08 21:55:02 +01:00
Michel Juillard
80ca47d62a
Merge remote-tracking branch 'local_master/master' into dr1break
2011-12-31 10:10:41 +01:00
Sébastien Villemot
30cbcdbb55
Store histval information for endogenous in M_.endo_histval
...
Really closes : #157
2011-12-28 11:35:22 +01:00
Sébastien Villemot
c541ceb849
Remove unused options_.deterministic_simulation_initialization
2011-12-28 10:44:34 +01:00
Stéphane Adjemian (Charybdis)
45d85f19fa
Added options for non linear filters + Cosmetic changes.
2011-12-26 17:46:48 +01:00
Stéphane Adjemian (Charybdis)
3e84e333b4
Changed verbosity options (fast deterministic simulations). Default is options_.verbosity==1.
2011-12-26 17:46:48 +01:00
Stéphane Adjemian (Charybdis)
e2f69ab4fc
Added arbitrary order of approximation for the Stochastic Extended Path approach.
...
Default is that we only consider shocks in t+1 to approximate the expectations. If
options_.ep.stochastic.order is set to s>1, shocks in t+1, t+2, ..., t+s are
considered. Obviously the (tensorial) tree of future shocks is growing
exponentially.
Other numerical rules of integration have to be added...
2011-12-23 17:57:41 +01:00
Houtan Bastani
87ffab3200
preprocessor: add symbol.options statement
2011-12-21 12:21:30 +01:00
Houtan Bastani
f5bfdbb23f
preprocessor: add prior statement
2011-12-21 12:21:29 +01:00
Houtan Bastani
1e78d70659
preprocessor: add data command
2011-12-21 12:21:29 +01:00
Houtan Bastani
29bead75c9
preprocessor: add set_time command
2011-12-21 12:21:29 +01:00
Michel Juillard
316c1b4dd6
Merge branch 'master' into dr1break
2011-12-16 20:50:18 +01:00
Stéphane Adjemian (Charybdis)
857eb7438f
Added provisions for future shocks (using Gaussian quadratures).
2011-12-16 11:42:03 +01:00
Michel Juillard
edd95a94c8
added options_.sub_draws in estimation for controlling the number of draws used in
...
computing the posterior distributions of various objects. Changed
options_.subdraws, used in the code, into options_.sub_draws.
2011-12-15 17:35:26 +01:00
Marco Ratto
b23c508671
Initialize new option analytic_derivation for estimation.
2011-12-13 17:05:55 +01:00
Stéphane Adjemian (Charybdis)
a36cf30118
Make options_.dynatol specific to x/f by adding dynatol.x and dynatol.f. By default dynatol.x=dynatol.f= old default value of dynatol.
2011-12-12 14:20:14 +01:00
Michel Juillard
7c62fafe0c
Merge branch 'master' into dr1break
2011-12-10 20:10:31 +01:00
Stéphane Adjemian (Charybdis)
6d9983df0c
New version of the extended path routines.
2011-12-09 18:03:30 +01:00