22956ea4a4
Removed useless instruction (default value of exit_flag is 1).
2012-11-21 10:11:30 +01:00
bb0993973a
Fixed output argument ordering in non_linear_dsge_likelihood.
...
Merge remote-tracking branch 'refs/remotes/fred/master'
2012-11-21 10:07:30 +01:00
2ee1ebbe47
fix a bug in the output order
2012-11-21 09:00:43 +01:00
db2615a4c7
Use economy-size decomposition option of qr2 routine.
2012-11-20 14:39:45 +01:00
af23d72cb1
Allow economy-size decomposition with qr2 routine.
2012-11-19 17:43:59 +01:00
9f1ad5568e
New nonlinear filters routines and bug fixes.
...
Merge remote-tracking branch 'refs/remotes/fred/master'
2012-11-19 16:23:07 +01:00
93e6b41df5
Added routines that will be used to import data (csv file) in a dynSeries object.
2012-11-19 16:17:00 +01:00
c3ff5d92d3
Cosmetic change.
2012-11-19 16:17:00 +01:00
5c893f501b
making extended path ready for parallel computing with parfor
2012-11-17 20:55:01 +01:00
c121aa14b1
Remove oo_.dr.{nstatic,npred,nboth,nfwrd,nspred,nsfwrd}
...
Replace them by equivalents in M_ (and an extra one: M_.dynamic).
IMPORTANT POINT: oo_.dr.npred used to count both purely backward and mixed/both
variables. This was the cause of lots of confusion. The new M_.npred only
counts purely backward variables.
We now have the following indentities:
M_.npred + M_.nboth + M_.nfwrd + M_.nstatic = M_.endo_nbr
M_.nspred = M_.npred + M_.nboth
M_.nsfwrd = M_.nfwrd + M_.nboth
M_.ndynamic = M_.npred + M_.nboth + M_.nfwrd
2012-11-16 20:05:13 +01:00
92727ed75c
Remove obsolete stoch_simul_sparse.m
2012-11-16 18:55:42 +01:00
ddbeb8ceaa
Remove unused oo_.dr.kae, oo_.dr.kad
2012-11-16 17:51:25 +01:00
31a13db66a
Remove duplicate information in M_.blocksMFS
2012-11-16 17:39:03 +01:00
6139af89d5
add pruning in the auxiliary particle filter.
2012-11-16 17:27:54 +01:00
bf19ab801f
Remove duplicate information in M_.blocksEQU
2012-11-16 17:16:40 +01:00
35c77c2e46
typo fixed.
2012-11-16 13:25:41 +01:00
915a203add
bug fixed.
2012-11-16 13:12:21 +01:00
919bd97682
add minor modifications.
2012-11-16 12:35:11 +01:00
202d8a2bd2
minor modifications
2012-11-16 12:35:11 +01:00
c8c7b9faf6
minor modifications.
2012-11-16 12:35:11 +01:00
eb9bc512d7
fixed bug in calculation of observed mean and variance + add the possibility for smooth resampling
2012-11-16 12:35:11 +01:00
3a9eed00df
fixed bugs for estimating a gaussian-mixture distribution.
2012-11-16 12:25:46 +01:00
186de044f4
fix bugs for sampling in a gaussian-mixture distributions.
2012-11-16 12:25:46 +01:00
f7e3ad7b25
fix bugs in k-means code.
2012-11-16 12:25:45 +01:00
f41e969dbe
computes the weights and the nodes for approximating gaussian distributions using the scaled unscented approach.
2012-11-16 12:25:45 +01:00
a184e80a89
computes the nodes and weights for approximating a gaussian distribution.
2012-11-16 12:25:45 +01:00
d5666529af
computes the density of particles in a gaussian-mixture distribution.
2012-11-16 12:25:45 +01:00
2fb1065b4f
computes the density of particles in a gaussian distribution.
2012-11-16 12:25:45 +01:00
56e723c08e
Computes the density of observables as a function of states.
2012-11-16 12:25:45 +01:00
eb72df7bc6
computes the proposal distribution during the importance sampling step for gaussian-mixture filters. Fix many bugs.
2012-11-16 12:25:45 +01:00
ab2ae38e8f
computes incremental weights for gaussian-mixture filters.
2012-11-16 12:25:45 +01:00
7dcdd20f6e
gaussian mixture nonlinear filters: uses gaussian-mixture approximations for particles. Fix bugs and normalize the way likelihood is calculated.
2012-11-16 12:25:45 +01:00
b0788fae50
computes the proposal distribution during the importance sampling step in gaussian nonlinear filters. Uses a nonlinear Kalman filter and several gaussian approximations.
2012-11-16 12:25:45 +01:00
553c26e0b3
gaussian nonlinear filter : uses a gaussian approximation for particles.
...
fix bugs and normalize the way we write the likelihood.
2012-11-16 12:25:45 +01:00
4725a2054d
compute incremental weights for gaussian filters.
2012-11-16 12:25:44 +01:00
a96850a090
fix bugs on sorting variables.
...
test different cases before doing resampling
2012-11-16 12:25:44 +01:00
f0fcfa3b62
fix bugs in the univariate smooth resampling approach.
2012-11-16 12:25:44 +01:00
c6e5368a9d
fix bug for weights initialization.
...
fix bug for calculating observed predicted mean and variance with the correct weights.
2012-11-16 12:25:44 +01:00
edf9955c09
fix bug when comparing xparam with upper and lower bounds.
2012-11-16 12:23:18 +01:00
b6c72f18fa
-add a variable to make a difference between gaussian approximations during importance sampling and approximation of states (for gaussian and gaussian-mixture filters)
...
-modify the default of parameters for the gaussian approximation using the scaled unscented transformation
2012-11-16 12:18:55 +01:00
6350962186
Fixed copyright.
2012-11-16 12:18:55 +01:00
7fd1df2505
Fixed bug when comparing xparam1 with upper and lower bounds.
2012-11-16 12:18:54 +01:00
79b5db3037
Cosmetic change.
2012-11-15 13:54:21 +01:00
89356ba248
Added a unitary test in the overloaded colon method of the dynDate class, using the new feature introduced in the previous commits.
2012-11-15 13:53:35 +01:00
67e8db365c
Changed subsref overloaded method (dynDate): added a new interface to populate empty dynDate objects with specified frequency. Added unitary test.
...
For instance if we instantiate an empty dynDate object for quaterly dates:
qq = dynDate('Q');
Then we can populate this object as follows (2009 second quarter):
d1 = qq(2009,2);
2012-11-15 12:40:24 +01:00
0b7e5910ae
Changed a condition in dynDate constructor (for yearly dates).
2012-11-15 12:34:11 +01:00
e71ccaef0b
Added the possibility to instantiate empty quaterly, monthly or weekly dynDate objects. Added unitary test.
2012-11-15 11:07:43 +01:00
2313ec2d94
Fixed function name, header and error message.
2012-11-14 12:32:03 +01:00
a1c8c785e0
Added unique method to the dynDates class.
2012-11-14 12:32:03 +01:00
bcaae72338
Added pop method to the dynDates class.
2012-11-14 12:32:02 +01:00
Stéphane Adjemian (Charybdis)
Frédéric Karamé
Michel Juillard
Sébastien Villemot