99060b0f80
gui: small improvements
2019-02-05 17:49:08 +01:00
fd09ba1fca
ols_style parsing: remove regressors that are equal to zero
2019-02-05 16:38:25 +01:00
b2fa7dd3c1
Merge branch 'discret_error_message' into 'master'
...
discretionary_policy_1.m: check for NaN before checking for non-zero derivatives
See merge request Dynare/dynare!1644
2019-02-05 14:42:33 +00:00
d5e25c8f43
mr_hessian.m: prevent infinite loop
...
Make sure hessian is not 0 and add counter.
Fixes #1636
2019-02-05 10:00:54 +01:00
8cb99bfe9d
discretionary_policy_1.m: check for NaN before checking for non-zero derivatives
...
Prevents false error message
2019-02-05 09:14:20 +01:00
5c75c30104
Fixed matlab warning.
...
[skip ci]
2019-02-04 16:26:48 +01:00
768555b523
Merge branch 'fixes_4.6' into 'master'
...
Fixes 4.6
See merge request Dynare/dynare!1641
2019-02-04 10:19:05 +00:00
b7c60ddf59
Added a routine for writing the problem to be solved to compute the BGP of a model.
...
- Only works with backward models.
- Probably doesn't work if the model includes auxiliary variables.
- Assumes that the trends are multiplicative.
2019-02-04 10:04:33 +01:00
553f1d3068
fix bug: line argument was being ignored
2019-02-01 17:05:10 +01:00
b2fcd86c35
print parts of model where there are evaluation errors
2019-02-01 17:05:10 +01:00
18b006d46f
fix comments
2019-02-01 16:45:13 +01:00
f29d974410
add copyright and fix spacing
2019-02-01 16:42:32 +01:00
c36db13bee
dseries: submodule update
2019-02-01 16:25:52 +01:00
31d59d6be3
-add file to read JSON gui perfect-foresight
2019-02-01 14:54:52 +01:00
eace56ce49
use switch statement instead of eval
2019-02-01 14:28:27 +01:00
8ddcccf18e
ols-style parsing: modify parsing to accommodate a larger set of expressions
2019-02-01 14:25:35 +01:00
c9423fe3ab
ols parsing: fix bug in check
2019-02-01 13:21:59 +01:00
ff2ca5fc17
remove unnecessary function
2019-02-01 12:04:46 +01:00
f4b8735431
pooled_ols: only save certain variables oo_ structure when passing back to pooled_fgls
2019-02-01 11:50:01 +01:00
0121e709fe
pooled_ols/pooled_fgls: fix bug in call to function
2019-02-01 11:45:16 +01:00
e758d5bdc9
ols parsing: fix bug: node value is not stored as string in AST
2019-02-01 11:37:41 +01:00
1f3e027a09
fix typo
2019-02-01 11:12:25 +01:00
a68f3fc082
remove unused output from functions
2019-01-31 16:11:35 +01:00
4c64b9108a
Revert "Revert "handle constant equations in preprocessor""
...
This reverts commit 850d1dcc95
2019-01-30 12:29:33 +01:00
c544f4a02c
small improvement
2019-01-30 12:29:08 +01:00
850d1dcc95
Revert "handle constant equations in preprocessor"
...
This reverts commit bcce997112
2019-01-29 00:38:01 +01:00
bcce997112
handle constant equations in preprocessor
2019-01-28 17:56:10 +01:00
d32de4deae
preprocessor, reporting: submodule update
2019-01-28 15:47:40 +01:00
cfc81bc5ff
Update preprocessor and dseries submodules
2019-01-28 15:16:37 +01:00
74520e77bf
Allow for lags on endogenous variable in growth option of pac_model.
2019-01-28 11:25:30 +01:00
668e641a6a
create report for SUR estimation
2019-01-25 18:11:27 +01:00
6d447f195a
sur: add model_name option
2019-01-25 18:11:26 +01:00
35cabed989
reporting: submodule update
2019-01-25 18:11:26 +01:00
d15c998804
reporting: submodule update
2019-01-25 18:02:16 +01:00
73ea0c2fe3
fix typo in comment
2019-01-24 12:42:08 +01:00
9bc1833a36
sur: fix bug in returned fitted value
2019-01-24 12:21:36 +01:00
ab5a19f08a
sur: fix bug in check of returned matrices
2019-01-24 11:38:40 +01:00
a402554083
fix bug in b219a99a16
2019-01-24 11:30:53 +01:00
b219a99a16
olsgibbs: fix bug in returned fitted value
2019-01-24 11:23:23 +01:00
d52ee5d290
fix comment
2019-01-24 11:20:39 +01:00
3690dc2e6f
Updated dseries submodule.
2019-01-23 17:26:08 +01:00
2482eafdb7
Fixed dseries class initialization in Octave.
2019-01-23 15:49:55 +01:00
99d770382b
change indexing into X.name
2019-01-23 15:48:46 +01:00
7632c4f071
use vobs instead of length(X.name)
2019-01-23 15:48:46 +01:00
f484942154
Update dseries submodule
2019-01-22 18:10:16 +01:00
d1afd56134
small fixes (comments, argument checks)
2019-01-22 16:11:03 +01:00
e1b01d9225
common parsing: jsonmodel not necessary as tag info is in AST
2019-01-22 16:10:42 +01:00
078e1ab99e
fix bug in ols common parsing routines causing ordering of regressor columns to change
2019-01-22 15:25:17 +01:00
9d413e2f8f
use vobs() instead of length(ds.name)
2019-01-22 15:23:42 +01:00
f514d443b6
aesthetic fix
2019-01-22 00:40:57 +01:00
63054a5728
olsgibbs: fix checks of input arguments
2019-01-22 00:08:09 +01:00
0406a2ae15
handle zeroed out equations in SUR
2019-01-21 15:57:11 +01:00
875381379f
sur: fix bug in creation of SUR matrices; handle random walk equation regardless of order in model equations
2019-01-18 15:07:24 +01:00
cf0d8f319e
remove extra whitespace
2019-01-18 14:09:40 +01:00
3293ff62c1
handle nested UnaryOpNodes in OLS-style equations
2019-01-18 11:50:10 +01:00
0badb00687
handle AR(1) with no parameter
2019-01-17 18:57:24 +01:00
cfd9eff6fb
simplify common parsing
2019-01-17 18:06:43 +01:00
b62ec7dc55
handle hard-coded constant*variable in AR(1)
2019-01-17 18:06:36 +01:00
a798889fe1
fix crashes of the expand interactive mode of shock decomp plots, requiring to store and use more info on the plot itself (manual cherry-pick from 2914805e0803165eb324081483b6e85e288c3ce0)
2019-01-17 11:37:41 +01:00
aeceafbec2
fixed error in formula of annualized steady state
...
(cherry picked from commit 696800b277dadd164244771ccfce8e39f1146472)
2019-01-17 10:57:39 +01:00
6e033a800b
trap possible cases where the option is type dates (manual cherry pick from commit 392ee4ae8a200a82003acd0c9b70c47c542f6b32)
2019-01-17 10:56:12 +01:00
47900a5909
trap slice sampler when doing mh_recover
...
(cherry picked from commit 4ba9c9296f1939fa14b1401886d9e9b86541e5e3)
2019-01-17 09:50:07 +00:00
beb165c002
bug fix with definition of steady state of plotted var when realtime is not zero
...
(manual cherry picked from commit 5badd34380f20e511e91b455482931c107a8f73b)
2019-01-17 10:47:39 +01:00
8713e46c4e
fixed bug with non-increasing time declarations of moment restrictions. Modified example to get more sensible graph of ACF.
...
(cherry picked from commit b0e7c47a000a679420970f1d05e4f2ed7d60764c)
2019-01-17 09:26:37 +00:00
4c133ac23a
olsgibbs: small fixes
2019-01-16 14:53:22 +01:00
556b5b6e88
small change to avoid an extra conditional
2019-01-16 14:53:22 +01:00
4dbbdb2c09
move common dates to parsing function
2019-01-16 14:53:22 +01:00
ee8e8c4282
small simplification
2019-01-16 14:53:22 +01:00
0fb88114c1
rename functions and change error message to be more general is parsing is shared between several different functions
2019-01-16 14:53:22 +01:00
bb9418a473
dyn_ols: only one output from function, remove varargout
2019-01-16 14:53:21 +01:00
03eaac8b6d
make plot_contributions use get_ast_jsonmodel
2019-01-16 14:53:21 +01:00
9a61c57102
remove functions that are no longer used
2019-01-16 14:53:21 +01:00
f755cfdcd4
surgibbs: small fixes
2019-01-16 14:53:21 +01:00
3247b1aa02
pooled_fgls: fix and update
2019-01-16 14:53:21 +01:00
e4dc7c403e
parse residual information
2019-01-16 14:53:21 +01:00
72081e640e
pooled_ols: fix dates
2019-01-16 14:53:20 +01:00
efe0a15e0e
comment out code
2019-01-16 14:53:20 +01:00
9ed97b88c1
account for potentially differing number of observations per equation when creating SUR matrices
2019-01-16 14:53:20 +01:00
a7181a8151
fix common parsing function to impose same dates per equation
2019-01-16 14:53:20 +01:00
3427d93c09
add comment to function
2019-01-16 14:53:20 +01:00
5d80fc903b
pooled_ols: fix up and use common parsing
2019-01-16 14:53:20 +01:00
1e41235b6b
common parsing: small fixes
2019-01-16 14:53:19 +01:00
934a22c3a7
sur: small fixes
2019-01-16 14:53:19 +01:00
ef391b1304
surgibbs: fix up and make work with sur changes
2019-01-16 14:53:19 +01:00
f8c0282b01
sur: fix up and use common code to create matrices
2019-01-16 14:53:19 +01:00
d8f26525b0
olsgibbs: fix up and use common parsing functions
2019-01-16 14:53:19 +01:00
a477075825
ols: further simplifications
2019-01-16 14:53:19 +01:00
a35fd291c5
ols: move json file parsing to a separate file
2019-01-16 14:53:19 +01:00
013124eedf
ols: move parsing to a helper function
2019-01-16 14:53:18 +01:00
093dd493d8
ols: catch unexpected case
2019-01-16 14:53:18 +01:00
7d35cb5c5a
fix typo
2019-01-14 15:02:32 +01:00
e8333421e7
remove unnecessary file
2019-01-10 10:44:19 +01:00
3bc6b34161
Bug fix.
...
Be agnostic about the name of the PAC model.
2019-01-07 13:33:08 +01:00
c94401ea66
remove spurious semicolon
2019-01-02 18:00:21 +01:00
42ad1a85ba
remove unnecessary find in statement
2019-01-02 17:56:45 +01:00
24eb3bcb34
remove unnecessary brackets from strings
2019-01-02 17:56:35 +01:00
6d5f86b464
Updated dseries submodule (fix for macOS).
...
X13 binary was not detected.
[ci-skip]
2018-12-27 10:59:38 +01:00
c7988c0717
Updated dseries submodule (moved dates class into dseries submodule).
2018-12-24 10:37:26 +01:00
f546eb67ed
Update preprocessor and dseries submodules
2018-12-20 17:10:22 +01:00
f363c3da6b
Restore behavior of 'nolog' option
...
The option had been broken by 0e77815ece
2018-12-20 17:10:22 +01:00
14305851a9
Updated dates submodule.
2018-12-19 17:24:00 +01:00
0e77815ece
Options list at the top of the mod-file now parsed by the preprocessor
...
Closes #1630
2018-12-19 16:15:48 +01:00
c829e7968d
If possible define boundaries for the share of optimizing agents in PAC/NLS.
...
The parameter is constrained to be in (0,1) for the following algorithms
- lsqnonlin (without Levenberg-Marquardt)
- fmincon
- annealing
- particleswarm
Other algorithms do not allow the definition of boundaries... We should add a
penality function to the objective for these algorithms.
2018-12-19 16:05:34 +01:00
df57bf7eb9
Fixed bug.
...
PAC/NLS routine was only working if the PAC model was named `pacman`.
2018-12-19 16:00:57 +01:00
11b24a0663
Add a constraint on the share of optimizing agents PAC/Iterative OLS.
...
Ensure that the parameter value lies betwwen 0 and 1.
2018-12-19 15:43:36 +01:00
c0570ff0af
Fixed typo.
2018-12-19 15:41:32 +01:00
23e37e45c4
Cosmetic change.
...
create a namespace for each {VAR/PAC} expectation model.
2018-12-19 11:43:26 +01:00
351da3b865
Do not update a dseries object, but instantiate a new object.
2018-12-19 10:53:09 +01:00
5db86aaf7f
Use global instead of local variable
2018-12-18 12:44:08 +01:00
fc74dded42
Increase tolerance in quantile.m unit test
...
The failure appeared because of (seemingly unrelated) changes in gamrnd, which
have modified the random seed when entering this test.
2018-12-17 12:38:41 +01:00
f2b5a909f3
Update dseries submodule.
2018-12-17 10:50:39 +01:00
6460a3a1e5
Document prior moments(distribution) option
2018-12-15 19:23:06 +01:00
68d3852cbf
Correctly update covariance matrix of exogenous shocks during prior simulation
2018-12-15 19:23:06 +01:00
b333f6bf2b
Correctly update covariance matrix of shocks for posterior draws
2018-12-15 19:23:05 +01:00
2ba9a8e10b
Prevent inconsistent specification in estim_params-block
...
Also documents the behavior
2018-12-15 18:49:53 +01:00
1880a34d2f
Provide error message if unsupported endogenous prior restriction is used
2018-12-15 18:44:53 +01:00
fec7455555
Make prior.m function account for endogenous prior restrictions
2018-12-15 18:44:53 +01:00
d5bdc19e19
Added unit tests for gamrnd algorithms.
...
Only tests first order moemnt, second order moment and Cumulative Distribution
Function (should also add some tests for independance).
2018-12-15 16:48:25 +01:00
e32b87d4b3
Streamlined gamrnd routine.
...
Give access to all the implemented algorithms through the third argument. The
last argument is a structure with two fields `large` and `small`. The first
field specifies the algorithm to be used for α>1 while the second field defines
the algorithm to be used for α ∈ (0,1).
Default algorithms are:
- Cheng (1977) for α>1,
- Johnk (1964) α ∈ (0,1).
2018-12-15 16:48:24 +01:00
d628b078ab
Fixed typo.
2018-12-15 16:48:24 +01:00
4206cabe52
Streamlined gamrnd algorithms.
2018-12-15 16:48:24 +01:00
8ac4860150
Fixed Weibull rejection algorithm.
...
The description given by Devroye (1986) is clearly wrong, this implementation
is based on the original paper Väduva (Series Statistics, 1977).
2018-12-15 16:48:24 +01:00
79ad4e952b
Bug fixes in Cheng (1977)'s gamrnd algorithm.
...
- Definition of Y was wrong (error in Devroye's book).
- Definition of X was wrong.
Also simplified the code w.r.t. update of the vector of indices.
2018-12-15 16:48:24 +01:00
162c8ab7d0
Set default value for the second hyperparameter.
2018-12-15 16:48:24 +01:00
426c63a735
Created namespace for gamrnd routines.
...
Also fixed implementation of Jonk's algorithm.
2018-12-15 16:48:23 +01:00
4775bca073
Fixed ordering of the columns of the Companion matrix...
...
For the level representation of the trend component model.
2018-12-14 17:37:35 +01:00
2cff485113
Fixed bug in the routine evaluating the (PAC/VAR) expectations.
2018-12-14 17:29:46 +01:00
38c8462142
If NLS, do not check the ordering of estimated parameters in PAC.
2018-12-12 16:28:19 +01:00
552c40cfc4
Update dseries and m-unit-tests submodules
2018-12-12 14:48:50 +01:00
139e922c0f
No longer downoald x13 binary within CI
...
The binary is now picked from the Debian package.
2018-12-12 11:06:20 +01:00
d2c10f216b
Update dseries submodule
2018-12-12 10:36:17 +01:00
8f5ba227d1
Updated dseries submodule.
2018-12-11 19:38:29 +01:00
89a8331eb6
Updated dseries submodule.
2018-12-10 23:00:52 +01:00
44e5a88ddd
dyn_ols: use AST instead of regex to parse equations
2018-12-07 19:54:32 +01:00
cb4384bb27
Create on the fly a routine for evaluating the (VAR/PAC) expectations.
...
The routine takes a dseries object as unique argument and return an updated
object with the expectation term.
If the mod file is named `example.mod` and if the (VAR/PAC) expectation model is
named `toto`, then after
var_expectation.print('toto');
the expectation term can be evaluated:
ts = example.var_expectations.evaluate_varexp(ts);
where ts is a dseries object containing all the time series appearign in the
auxiliary (var or trend_component).
2018-12-03 15:07:43 +01:00
53ab321de8
Added the possibility to use linear expressions in VAR_EXPECTATION_MODEL.
...
See tests/var-expectations/9/example.mod for a self documented example.
Also updated all the integration tests using the option `expression` instead of
`variable` (which is deprecated and will be removed at some point).
2018-11-30 23:36:06 +01:00
650c78f99d
Merge branch 'master' into ecb-master
2018-11-30 21:34:56 +01:00
d501d6d511
Added interface to lsqnonlin (Mathworks' optimization toolbox) in pac.estimate.nls.
2018-11-29 10:29:55 +01:00
c79be57447
Fixed definition of h (PAC expectation reduced form).
2018-11-27 10:05:40 +01:00
e6c716ae9b
Added the possibility to use Gauss-Newton in pac/nls.
...
Also added the computation of the covariance matrix of the NLS estimator (using
White and Domovitz approach) and integration test.
2018-11-26 09:53:18 +01:00
139c58dd76
Added Gauss Newton algorithm for minimizing sum of squared residuals.
2018-11-26 09:48:50 +01:00
efec664667
Preprocessor update + corresponding change in parameters derivatives
...
In the *_params_derivs.m files, in the "hp" matrix, the preprocessor now fully
outputs the symmetric elements, so getH needs to be adapted.
By the way, also clean up code for "rpp" and "gpp", were full symmetric
elements were also already present (and therefore do not need to be duplicated
in getH).
2018-11-22 17:53:39 +01:00
cf35496c06
Fixed bug in the evaluation of the sum of squared residuals (PAC/NLS).
2018-11-21 15:21:23 +01:00
7d2d0d6590
Added the possibility to estimate the share of optimizing agents by iterative OLS (PAC).
2018-11-21 15:07:15 +01:00
635aa65665
Merge branch 'ident_plot_bug' into 'master'
...
identification_analysis.m: Fix bug introduced in 60d95b65f2Dynare/dynare!1637
2018-11-16 16:21:14 +00:00
b615d2d795
identification_analysis.m: Fix bug introduced in 60d95b65f2
...
We already used a variable called offset
2018-11-16 16:20:27 +01:00
4ecbcbb136
evaluate_steady_state_file.m: Remove redundant setting of auxiliary variables
...
The original problems seems to have been fixed by #1133
Essentially reverts 3c7e60b744closes #1175
2018-11-16 14:12:55 +01:00
f947a31b80
evaluate_steady_state_file.m: fix two bugs in debugging mode
2018-11-16 09:24:43 +00:00
60d95b65f2
Identification strength: make plotting consistent
...
1. Rely on actually computed standard deviations in bayestopt_ instead of potentially unset estim_params_
2. Remove arbitrary normalizations/omitted normalization in case of division by 0 in normalization
3. Distinguish between 0 identification and division by 0 due to normalization in plots
(cherry picked from commit 7341e21a381850c47fbed018bf6a7acdda4fa92e)
2018-11-16 10:17:30 +01:00
ad451ad224
Merge branch 'prior_dens_info' into 'master'
...
Deal with info returned by priordens being a vector
See merge request Dynare/dynare!1629
2018-11-15 09:39:11 +00:00
513f3ec68e
Merge branch 'evaluate_steady_state' into 'master'
...
evaluate_steady_state.m: consistently use params
Closes #1627
See merge request Dynare/dynare!1630
2018-11-15 09:22:09 +00:00
30361ccb49
GSA scatter plots: Correctly pass LaTeX-names to figure
2018-11-14 18:37:54 +01:00
022913579a
map_calibration.m: Fix bug that prevented display of restrictions
2018-11-14 18:37:05 +01:00
c16c2fb894
evaluate_steady_state.m: consistently use params
...
Prevents erroneously relying on stale parameters from M.params (thanks to Tom Holden); closes #1627
2018-11-14 16:01:55 +01:00
09a166f4e7
Revert "adding missing any()"
...
This reverts commit a5b97a28a0
2018-11-14 15:04:27 +01:00
c1b0bb5ee3
initial_estimation_checks.m: Turn vector info into scalar
...
Related to a5b97a28a0
2018-11-14 15:02:15 +01:00
18faa9733c
Further updates related to minimal MATLAB version bump
2018-11-13 18:29:20 +01:00
7a75872f72
Modernization: use tilde (~) syntax for ignored output arguments
2018-11-13 18:02:09 +01:00
ce49cd95df
Add missing iscolumn and ismatrix for MATLAB < R2010b
...
Also fix a bug in isrow (it was not checking the rank of the array).
2018-11-13 16:59:27 +01:00
736d43c4e6
Bump minimal MATLAB version requirement to 7.9 (R2009b)
2018-11-13 15:59:16 +01:00
096936be9c
The preprocessor now compiles the MEX when use_dll is specified
...
Update documentation accordingly (except on macOS where the solution has not
yet been implemented).
Also include MinGW in the Windows installer.
2018-10-29 14:41:34 +01:00
cc4183d3a5
Fix for recent MATLABs under Windows
2018-10-26 15:02:26 +02:00
8cb51e26e6
options_.instruments is now a cell array
...
This fixes crashes under Octave (while MATLAB seems to accept array-indexing
with cell arrays).
2018-10-25 15:01:53 +02:00
dcea7514d9
Merge branch 'master' into ecb-master
2018-10-24 18:31:31 +02:00
8a2c38cf6c
olsgibbs: save fitted values in dataset; allow use of dictionary as in dyn_ols
2018-10-24 17:17:01 +02:00
510bc2e299
Removed unused variable.
2018-10-24 12:08:00 +02:00
1c1a3ea0a4
Removed tests on targets.
...
A target does not need to be a random walk. Also added a new integration
test (where the target is a constant, 0).
2018-10-24 12:06:14 +02:00
a53c63d6d5
Added routines to print PAC_EXPECTATION and VAR_EXPECTATION in files.
...
Each new routine pac.print() and var_expectation.print(), creates two files
that can be included (using the macro language) in a mod file. The first
file, {pac,var}expectationmodename-parameters.inc saved under
{M_.fname}/model/{pac,var}expectationmodel, contains the declaration of the
parameters created for the expanded (var or pac) expectation model, with
calibrated values. The second file, {pac,var}expectationmodename-parameters.inc
saved in the same subfolder, contains the expanded version of the (pac or var)
expectation term as a linear combination of the variables in the companion
representation of the expectatino model.
2018-10-14 16:57:51 +02:00
e5317b17de
Cosmetic change.
2018-10-06 17:52:58 +02:00
108854a484
Bug fix. verbosity option was not honoured in simplex algorithm (8).
...
Thanks to Reuben for reporting.
2018-10-06 17:50:37 +02:00
dfafe86246
Save teh posterior mean of the slope parameters as a column vector.
2018-10-06 16:55:42 +02:00
82187f9d41
Fixed olsgibbs routine.
2018-10-06 16:08:28 +02:00
798aaaa968
Cosmetic change.
2018-10-04 23:17:49 +02:00
ee2ad1ee15
olsgibbs: linear regression with Gibbs sampler
2018-10-04 17:12:47 +02:00
57bf291c00
add comment
2018-10-04 17:12:47 +02:00
ae71599292
Switch to new get_companion_matrix routine.
2018-09-28 19:27:17 +02:00
bbbaabc702
Cosmetic change.
2018-09-28 12:26:59 +02:00
3ad654b9b8
Allowing for different number of periods in conditional_forecast_path.
...
Updated preprocessor submodule.
2018-09-27 18:58:24 +02:00
07141a8681
Various optimizations.
2018-09-26 22:51:05 +02:00
7c83ba7ea7
Fixed documentation header.
2018-09-26 22:51:05 +02:00
973c3955f0
adjust for ‘targets’ keyword instead of ‘trends’
2018-09-25 14:58:36 +02:00
974bcd92ae
Merge branch 'ecb-master' into ecb-master-targets
2018-09-25 14:41:11 +02:00
c6d5b09c29
fix typo
2018-09-25 14:39:48 +02:00
b3a8062b46
Added missing output.
2018-09-19 17:39:58 +02:00
031e2c87c6
New version of get_companion_matrix + new integration test.
...
The routine is still buggy. In PAC and VAR_EXPECTATION models we
use get_companion_matrix_legacy routine instead.
2018-09-19 17:39:48 +02:00
2cf6ccdaa5
Updated dseries submodule.
2018-09-18 17:37:37 +02:00
3e6e448134
Updated dates and dseries submodules.
2018-09-14 23:13:13 +02:00
99aadf7d1a
Updated m-unit-tests submodule.
2018-09-14 15:37:30 +02:00
62301fa1e9
dyn_ols: fix typo
2018-09-14 15:09:31 +02:00
2518cd9884
Raise error of an optimization algorithm is not available.
2018-09-13 16:16:07 +02:00
2437d7034e
Give access to more than one optimizer in PAC estimation (NLS).
...
Default is to use csminwel optimizer (because it seems to do the job and does
not rely on the Mathworks toolbox).
2018-09-13 16:15:52 +02:00
eb316d7e2d
update files for change of option name from `trends` to `targets`
2018-09-13 12:24:32 +02:00
5ab8120780
get companion matrix using preprocessor output instead of get_ar_ec_matrices.m
2018-09-12 18:38:26 +02:00
74094683de
get_companion_matrix: fix header
2018-09-12 18:29:21 +02:00
a6150bae82
get_companion_matrix: simplify check (no need for allocation)
2018-09-12 18:29:14 +02:00
a32e2f165a
get_companion_matrix: remove unnecessary conditional
2018-09-12 18:29:07 +02:00
e9d7f6ca24
Merge branch 'master' into ecb-master
2018-09-12 18:28:47 +02:00
6815624a6a
Fixed bug.
...
Growth neutrality correction was applied twice.
2018-09-12 15:41:07 +02:00
51416bf2ae
Always use "isoctave" for testing Octave context
2018-09-07 17:22:08 +02:00
0f3678ec4a
Fixed ordering of variables in error correction term of PAC equation
...
PAC equation has to be written as
diff(x) = a0*(xstar(-1)-x(-1)) + a1*diff(x(-1)) + ... + ap*diff(x(-p)) + PAC_EXPECTATION(pacmodelname) + ...;
In the error correction term, a0*(xstar(-1)-x(-1)), we must have the difference
between the target (the trend xstar(-1)) and the level of the endogenous
variable (x(-1)). To ensure stability around the trend, the parameter a0 needs
to be positive.
REMARKS
[1] In the TREND_COMPONENT_MODEL the error correction terms are written in
reverse order, ie as the difference betwwen the level of the endogenous
variable and the trend variable.
[2] In the estimation routine we do not constrain a0 to be positive, but is
would surely help to satisfy this condition in the initial condition.
2018-09-06 22:07:31 +02:00
74ef4d2131
Cosmetic change.
2018-09-06 13:40:43 +02:00
5d8a1feda4
Fixed regular expression.
...
Data matrices were not correcty lagged for lags>1. This commit fixes the
inconsistencies (in terms of residuals) between Iterative OLS and NLS routines.
2018-09-06 13:40:27 +02:00
940c970ff3
Use same range in NLS and Iterative OLS.
...
The range option is the range of data considered for the right hand side variable.
2018-09-04 09:40:47 +02:00
6e1d06bddc
Display SSR in Iterative OLS.
2018-09-03 18:01:59 +02:00
8b1b1273c3
Fixed correction for growth neutrality.
...
Also added the possibility to use an endogenous variable for growth.
2018-09-03 17:57:10 +02:00
fde103d5f4
Removed debugging display.
2018-09-03 17:55:55 +02:00
a207550680
Fixed error correction term.
2018-09-03 17:53:01 +02:00
044f0f8de6
Added missing output required by some integration tests.
2018-09-03 17:38:09 +02:00
5ea0c8098b
Ensure that the ordering of the trend variables is consistent...
...
... With the ordering of the EC variables.
2018-09-03 17:35:59 +02:00
e1b2b80d64
Small efficiency change.
2018-08-29 17:42:49 +02:00
d23d1a470b
Added the possibility to estimate PAC_EQUATION with iterative ols and non constant growth.
2018-08-29 17:42:04 +02:00
30b34c4323
Small efficiency changes.
2018-08-29 16:15:52 +02:00
00498d19c8
Fixed ec matrix (rules out lag>1).
2018-08-28 19:25:05 +02:00
36388ce28a
Fixed bug in PAC_EXPECTATION.
...
The selection of the target variable in the error correction term was still
wrong if the target variable was not an auxiliary variable.
2018-08-28 19:23:35 +02:00
18a2626543
Fixed bugs in PAC_EXPECTATION.
...
- The selection of the target variable in the error correction term was wrong.
- Target variables in logs were still not working.
2018-08-28 12:12:21 +02:00
6dc35c676b
Fixed bug in VAR_EXPECTATION with TREND_COMPONENT_MODEL.
...
The selection of the expected variable in the companion VAR was not correct.
2018-08-28 11:06:20 +02:00
ae530293c5
Fixed bug. Trends in log where not accounted for in PAC equation.
2018-08-27 19:35:33 +02:00
83c9b0d5b6
Make rool for trend_component_model in var_expectation.
...
Not yet working, a bug in the preprocessor remains to be fixed. The
preprocessor does not create the correct number of reduced form parameters
for VAR_EXPECTATION when the auxiliary model is a trend component model,
because it ignores the fact that the model has to be rewritten in levels.
2018-08-27 14:37:26 +02:00
6ec9758313
update var_expectation code and tests for change in field name
2018-08-23 10:38:39 +02:00
cd05b9ba8f
Cosmetic changes.
2018-08-22 11:49:30 +02:00
2b6ee881a4
Efficiency changes.
2018-08-22 11:49:15 +02:00
442a70e905
Added missing semicolon.
2018-08-21 21:56:01 +02:00
4a0b3ffe7e
Adapted matlab code to new TREND_COMPONENT auxiliary model.
2018-08-21 21:49:58 +02:00
38b2593e31
Efficiency changes.
2018-08-21 21:02:20 +02:00
322682635f
get_ar_ec_matrices: fix bug in ordering of ec matrix
2018-08-20 14:58:52 +02:00
b27267da9b
get_ar_ec_matrices: changes for introduction of trend_component_model statement
2018-08-20 10:57:26 +02:00
7d6296f7a6
Fixed bug introduced by 42bba08887.
2018-08-17 21:42:09 +02:00
e50bb1566e
Updated dseries submodule.
2018-08-16 16:42:25 +02:00
822446c043
Added missing copyright headers.
2018-08-15 22:05:47 +02:00
42bba08887
Make the code robust to the order of declaration of the estimated parameters.
2018-08-15 22:03:33 +02:00
15ad3cebcf
Fixed indexation bug (last autoregressive parameter was missing).
2018-08-15 22:03:01 +02:00
702e6d4cc7
Fixed lag number in PAC equation.
...
The number of autoregressive parameters has to be equal to the number of lags
in the PAC equation.
2018-08-13 12:41:37 +02:00
8b3ba814c7
Fixed PAC index for target variable in companion VAR.
2018-08-10 15:42:13 +02:00
8a6d23f471
Merge branch 'master' into ecb-master
2018-08-10 15:37:26 +02:00
451385774b
Cosmetic changes.
2018-08-10 15:36:41 +02:00
86445832fd
Added routine returning true iff a variable is an auxiliary variable.
2018-08-10 14:52:22 +02:00
c28c8187e8
Fixed typo.
2018-08-08 16:23:15 +02:00
8033b2737e
Added routines for VAR_EXPECTATION_MODEL.
2018-08-08 15:53:55 +02:00
c9a38c4de7
Merge branch 'master' into ecb-master
...
Fixed conflicts:
- tests/run_block_byte_tests_octave.m
2018-08-04 21:08:08 +02:00
5b30ed34ae
Allow calls to get_companion_matrix routine with only one argument.
2018-08-04 20:15:52 +02:00
dc026329a7
Cosmetic change.
2018-08-04 20:14:28 +02:00
13526bcbce
Octave now has the same graphical waitbar as MATLAB (for MCMC progress)
2018-08-02 15:33:18 +02:00
d03cc56587
solve_algo=7 and stack_solve_algo=2 are now allowed under Octave
...
The gmres() function has been implemented (actually since Octave 3.4.1!).
2018-08-02 15:20:39 +02:00
fda021855c
Fix forecast command with deterministic exogenous and linear option
...
Closes #1608
2018-08-02 14:20:51 +02:00
f830fb7d53
Updated dseries submodule.
2018-07-31 12:55:17 +02:00
ecb41ef4c0
Merge branch 'master' into ecb-master
2018-07-28 10:54:27 +02:00
6d79ca4f58
Return 0 if input is empty.
2018-07-28 10:48:43 +02:00
a5b97a28a0
adding missing any()
2018-07-27 20:24:13 +02:00
a144aad150
Merge branch 'master' into ecb-master
2018-07-27 09:47:22 +02:00
433a68169d
TaRB: fix bug where incorrect last posterior was returned if last draw was rejected
2018-07-24 17:07:29 +02:00
157641ac91
Updated particles submodule.
2018-07-24 14:12:53 +02:00
7d7449c775
Merge branch 'master' into ecb-master
...
Fixed conflicts:
matlab/modules/dseries
2018-07-23 17:05:30 +02:00
0edebbe5d7
Added Iterative OLS routine for PAC equation.
2018-07-23 16:12:46 +02:00
cffff796f1
Added routine to test if an auxiliary variable is a lag of a variable.
2018-07-23 10:41:34 +02:00
f03a9b98e1
Fixed typo.
2018-07-20 17:09:15 +02:00
7421e6b5b2
Fixed copyright header.
2018-07-19 15:36:02 +02:00
44c49eaef8
Improved iterations display of Dynare's simplex routine.
2018-07-19 15:15:05 +02:00
2d321cb0e3
Increased default number of iterations in simplex.
2018-07-19 15:15:05 +02:00
97229177b4
simplex_optimization_routine.m: account for penalty indicator being third output, not fourth one
...
Caused occasional crashes
2018-07-19 09:25:27 +02:00
5361325ba9
Fixed unit test (false negative in Octave testsuite).
2018-07-18 12:18:43 +02:00
ac3af2052a
Remove test from Octave testsuite.
...
The function is only called by Matlab if the statistics toolbox is not available.
2018-07-18 11:14:47 +02:00
1d2d7e0776
Updated dseries submodule (and matlab-fame-io).
2018-07-18 10:42:55 +02:00
064b270795
Updated dseries submodule (Octave testsuite fixes).
2018-07-17 18:41:25 +02:00
8b0d673458
Cosmetic change (Octave unit test).
...
(cherry picked from commit 48ee13646a072d0ad9466f901f2d048594717d71)
2018-07-17 17:44:42 +02:00
0d27bd8dea
Fixed unit tests on Weibull pdf (Octave testsuite).
...
(cherry picked from commit bc962a91c0749e6c9f108c867acde134e6bb80fc)
2018-07-17 17:44:15 +02:00
4154817acc
Added missin semicolons.
...
(cherry picked from commit 4a0b1c2e93d23c85a7874b1bbea85819f56739f5)
2018-07-17 17:43:58 +02:00
9d01b17e03
Removed one unit test from Octave testsuite.
2018-07-17 11:30:20 +02:00
cd9410644f
Moved PAC estimation routine (NLS).
2018-07-16 16:30:07 +02:00
9468c5b61d
Bug fix in PAC equation simulation/estimation.
...
Fixed the definition of idns (must target the trend variable).
2018-07-16 15:38:04 +02:00
c46fe7fcfa
Added check routine for PAC equation.
2018-07-16 14:56:05 +02:00
0074561182
Updated dseries submodule (unit test fix).
2018-07-14 11:54:01 +02:00
2256acfa73
Fixed replacement of LHS variable if the equation has no lag.
2018-07-13 19:10:25 +02:00
9b51b1ee16
Fixed logic for replacing variables by views of the data array.
2018-07-13 19:07:38 +02:00
7685a0ccda
Octave compatibility fix
...
Now that the set_auxiliary_variables file is in a "+" directory, Octave fails
at str2func if the file does not exist.
2018-07-13 17:22:27 +02:00
7e7f9b2e2e
Updated dseries submodule.
2018-07-13 14:04:57 +02:00
710158ba3e
Updated dseries submodule.
2018-07-12 19:50:53 +02:00
30c41733bc
Updated dseries submodule.
2018-07-12 19:38:15 +02:00
d7e0c5b868
Merge branch 'master' into ecb-master+folder
2018-07-11 11:08:04 +02:00
ebad639546
Bug fix (parenthesis mismatch).
2018-07-11 11:03:15 +02:00
a31611a89b
Adapted codes to the json files new location.
...
(cherry picked from commit f0a5a2aac9
2018-07-10 20:19:43 +02:00
f0a5a2aac9
Adapted codes to the json files new location.
2018-07-10 20:17:43 +02:00
c5ebb2ca30
Adapted codes to the json files new location.
2018-07-10 20:15:53 +02:00
24e2a76c39
Remove condition.
...
The checkdatabase routine can be used on models with leads.
2018-07-10 18:50:15 +02:00
51332c6fd6
Fixes related to the new "+" subfolder.
2018-07-10 18:34:49 +02:00
85a585c73e
Merge branch 'master' into ecb-master+folder
...
Fixed conflicts in:
tests/block_bytecode/run_ls2003.m
2018-07-10 10:18:55 +02:00
e50d4fd41c
Added routine to update parameter values in base workspace.
2018-07-09 18:42:03 +02:00
bc54022da6
Fixed row ordering issue in the trend component model.
2018-07-09 18:41:35 +02:00
829849f4db
Fixed regular expression used to identify a PAC auxiliary model.
2018-07-05 11:50:50 +02:00
4a9c752fd1
Fixed the definition of a lag polynomial.
2018-07-04 14:57:54 +02:00
3863a74319
Fix bug introduced in 13ce91861 in graph decomposition under MATLAB
2018-07-04 12:33:51 +02:00
227ac46aa9
Fixed bug related to the parameters in PAC equation.
2018-07-04 10:02:19 +02:00
f248feb839
Fixed function handle name for sum of square residuals.
2018-07-04 09:38:19 +02:00
33d752bad3
Fixed filtering out of non exact matches.
...
Patch proposed by Anatole, see:
https://gitlab.com/McModelTeam/McModelProject/issues/76
2018-07-04 08:57:07 +02:00
13ce918612
Octave compatibility fixes in graph decomposition
2018-07-02 16:06:11 +02:00
b2c554623f
Various fixes related to the new layout of generated files
2018-06-28 18:14:40 +02:00
87bb0c6997
Fix mistake in previous commit w.r.t. steadystate file
2018-06-27 17:15:13 +02:00
a1b8bd39b2
Move the location of various generated files on the filesystem
...
- M and MEX files are now under +${MODELNAME}/
- bytecode, C source and JSON now under ${MODELNAME}/model/
2018-06-27 17:03:39 +02:00
37bd4a1fea
Added routine for estimating a PAC equation by NLS.
...
This is not a final version. The routine does not print the estimation result, and does not
compute the variance of the estimator. The routine updates the content of M_.params with the
vector of PAC parameters minimizing the sum of squared residuals.
2018-06-21 18:48:58 +02:00
8d1664898e
Merge remote-tracking branch 'github/master' into ecb-master
2018-06-21 18:48:16 +02:00
45115e1172
Replaced pac.update.equation() by pac.update.expectation().
2018-06-21 18:47:57 +02:00
Houtan Bastani
Sébastien Villemot
Johannes Pfeifer
Stéphane Adjemia (Scylla)
Sumudu Kankanamge
Marco Ratto
Marco Ratto
Stéphane Adjemian (Charybdis)
Michel Juillard
Stéphane Adjemian(Charybdis)
Michel Juillard