June 2016 - Dev - dynare.org

Status of testsuite in master branch
by Dynare Robot 30 Jun '16

30 Jun '16

848d81eb84ae77165d00d5841a3045501b380cf5 Merge pull request #1238 from rattoma/gsa ================================ DYNARE MAKE CHECK MATLAB RESULTS ================================ | TOTAL: 692 | PASS: 681 | FAIL: 5 | XFAIL: 6 | XPASS: 0 | LIST OF FAILED TESTS: | * kalman/likelihood_from_dynare/fs2000_corr_ME.mod | * kalman/likelihood_from_dynare/fs2000_corr_ME_missing.mod | * lmmcp/sw_newton.mod | * observation_trends_and_prefiltering/calib_smoother/Tr_no_prefilter_loglin_calib_smoother.mod | * observation_trends_and_prefiltering/calib_smoother/Tr_prefilter_loglin_calib_smoother.mod | | LIST OF 10 SLOWEST TESTS: | * walsh.mod - 1268.620000 | * optimizers/fs2000_6.mod - 860.830000 | * moments/example1_hp_test.mod - 849.590000 | * ep/rbc_mc.mod - 454.370000 | * estimation/TaRB/fs2000_tarb.mod - 437.460000 | * observation_trends_and_prefiltering/MCMC/Trend_loglin_no_prefilt_first_obs_MC.mod - 427.750000 | * observation_trends_and_prefiltering/MCMC/Trend_no_prefilter_first_obs_MC.mod - 426.950000 | * observation_trends_and_prefiltering/MCMC/Trend_no_prefilter_MC.mod - 419.520000 | * observation_trends_and_prefiltering/MCMC/Trend_prefilter_MC.mod - 415.930000 | * observation_trends_and_prefiltering/MCMC/Trend_loglinear_no_prefilter_MC.mod - 415.570000 A full log can be found at http://www.dynare.org/testsuite/master/matlab

1 0

Status of testsuite in master branch
by Dynare Robot 27 Jun '16

27 Jun '16

c5f8fe7f852dc0451ec98d3d598cdf5a814d5c5e Added integration test for {variables,parameters} partitioning. ================================ DYNARE MAKE CHECK MATLAB RESULTS ================================ | TOTAL: 692 | PASS: 681 | FAIL: 5 | XFAIL: 6 | XPASS: 0 | LIST OF FAILED TESTS: | * kalman/likelihood_from_dynare/fs2000_corr_ME.mod | * kalman/likelihood_from_dynare/fs2000_corr_ME_missing.mod | * lmmcp/sw_newton.mod | * observation_trends_and_prefiltering/calib_smoother/Tr_no_prefilter_loglin_calib_smoother.mod | * observation_trends_and_prefiltering/calib_smoother/Tr_no_prefil_f_obs_loglin_cal_smoother.mod | | LIST OF 10 SLOWEST TESTS: | * walsh.mod - 1240.790000 | * optimizers/fs2000_6.mod - 862.710000 | * moments/example1_hp_test.mod - 831.490000 | * ep/rbc_mc.mod - 457.590000 | * estimation/TaRB/fs2000_tarb.mod - 434.040000 | * observation_trends_and_prefiltering/MCMC/Trend_loglin_no_prefilt_first_obs_MC.mod - 426.760000 | * observation_trends_and_prefiltering/MCMC/Trend_no_prefilter_first_obs_MC.mod - 424.640000 | * observation_trends_and_prefiltering/MCMC/Trend_no_prefilter_MC.mod - 417.200000 | * observation_trends_and_prefiltering/MCMC/Trend_loglinear_no_prefilter_MC.mod - 414.920000 | * observation_trends_and_prefiltering/MCMC/Trend_prefilter_MC.mod - 402.240000 A full log can be found at http://www.dynare.org/testsuite/master/matlab

1 0

Missing SGU mod-file from Wiki
by Johannes Pfeifer 27 Jun '16

27 Jun '16

Hi, the user at http://www.dynare.org/phpBB3/viewtopic.php?f=2&t=8319 is looking for the model_sgu07_welf.mod at http://www.dynare.org/DynareWiki/OrderTwo Does anyone know abotu the wherabouts of that file? We also might want to update that wiki-site. Best, Johannes

3 5

Status of testsuite in master branch
by Dynare Robot 26 Jun '16

26 Jun '16

581198741f357d56a29bb8e875afc01848b4e2bf Revert "Revert "preprocessor: allow for partitioning of variables along multiple dimensions"" ================================ DYNARE MAKE CHECK MATLAB RESULTS ================================ | TOTAL: 691 | PASS: 679 | FAIL: 6 | XFAIL: 6 | XPASS: 0 | LIST OF FAILED TESTS: | * kalman/likelihood_from_dynare/fs2000_corr_ME.mod | * kalman/likelihood_from_dynare/fs2000_corr_ME_missing.mod | * lmmcp/sw_newton.mod | * observation_trends_and_prefiltering/ML/Trend_no_prefilter_first_obs.mod | * observation_trends_and_prefiltering/calib_smoother/Tr_no_prefilter_loglin_calib_smoother.mod | * observation_trends_and_prefiltering/calib_smoother/Tr_no_prefil_f_obs_loglin_cal_smoother.mod | | LIST OF 10 SLOWEST TESTS: | * walsh.mod - 1072.960000 | * optimizers/fs2000_6.mod - 863.490000 | * moments/example1_hp_test.mod - 845.360000 | * ep/rbc_mc.mod - 455.180000 | * estimation/TaRB/fs2000_tarb.mod - 436.410000 | * observation_trends_and_prefiltering/MCMC/Trend_loglin_no_prefilt_first_obs_MC.mod - 426.160000 | * observation_trends_and_prefiltering/MCMC/Trend_no_prefilter_MC.mod - 425.350000 | * observation_trends_and_prefiltering/MCMC/Trend_no_prefilter_first_obs_MC.mod - 424.770000 | * observation_trends_and_prefiltering/MCMC/Trend_loglinear_no_prefilter_MC.mod - 419.400000 | * observation_trends_and_prefiltering/MCMC/Trend_prefilter_first_obs_MC.mod - 401.680000 A full log can be found at http://www.dynare.org/testsuite/master/matlab

1 0

Kalman filter
by Johannes Pfeifer 18 Jun '16

18 Jun '16

Hi Marco (or anyone who can answer the questions), two quick questions: 1. Is there a reason https://github.com/DynareTeam/dynare/pull/1088/commits/05fc096569e15e89d8d13 b08799321c0313b168d was not ported to the diffuse filter? 2. What is the logic behind https://github.com/DynareTeam/dynare/pull/1088/commits/93e7be66e7b08bb0fd46b e7863dd9834c3baafb5 where diffuse_kalman_tol is used for evaluating the singularity of Finf, but Kalman_tol for Fstar? Isn't that pretty much the same check for singularity of the same matrix? diffuse_kalman_tol also governs the singularity Pinf which is the criterion to leave the diffuse filter. 3. What is the reference for if any(diag(F)<kalman_tol) || rcond(F./(sig*sig')) < kalman_tol if ~all(abs(F(:))<kalman_tol) return else a = T*a; P = T*P*transpose(T)+QQ; end If F is singular we turn to the univariate filter if F is 0 everywhere. If F is not 0 everywhere, we continue with a Kalman gain of 0 and discard the observation from the likelihood. But the gain in this case should be more like K=P*Z'/F= P*Z'/0=Inf. Shouldn't we generally turn to the univariate filter in this case instead of discarding the observation? Best, Johannes ------ Johannes Pfeifer Department of Economics University of Mannheim L7, 3-5, Room 242 68131 Mannheim Germany +49 (0)621 181-3430

3 4

missing matlab/missing/econ folder
by Stéphane Adjemian 16 Jun '16

16 Jun '16

Hi Marco (and Johannes), I do not know when this problem appeared exactly (a couple of weeks probably), but when I run Dynare I obtain this warning: Warning: Name is nonexistent or not a directory: /home/stepan/works/dynare/matlab/missing/econ > In path (line 109) In addpath (line 88) In dynare_config (line 88) In dynare (line 63) Using git blame on matlab/dynare_config, I see that the addpath for missing/econ was introduced by you in commit #22f51d1e75854ae6db9f748217965a69d90a534d, but I do not find the targeted folder (which should contain autocorr)... Any clue on this? Best, Stéphane. -- Stéphane Adjemian Université du Maine, GAINS & CEPREMAP

3 2

Status of testsuite in master branch
by Dynare Robot 16 Jun '16

16 Jun '16

33d35bc3c5d81b0dc57f4a0b4df92ad82d036aba Added quantile routine in missing. ================================ DYNARE MAKE CHECK MATLAB RESULTS ================================ | TOTAL: 691 | PASS: 682 | FAIL: 3 | XFAIL: 6 | XPASS: 0 | LIST OF FAILED TESTS: | * kalman/likelihood_from_dynare/fs2000_corr_ME.mod | * kalman/likelihood_from_dynare/fs2000_corr_ME_missing.mod | * lmmcp/sw_newton.mod | | LIST OF 10 SLOWEST TESTS: | * walsh.mod - 1060.620000 | * optimizers/fs2000_6.mod - 866.130000 | * moments/example1_hp_test.mod - 850.060000 | * ep/rbc_mc.mod - 457.180000 | * estimation/TaRB/fs2000_tarb.mod - 439.570000 | * observation_trends_and_prefiltering/MCMC/Trend_prefilter_MC.mod - 437.700000 | * observation_trends_and_prefiltering/MCMC/Trend_loglin_no_prefilt_first_obs_MC.mod - 430.730000 | * observation_trends_and_prefiltering/MCMC/Trend_no_prefilter_first_obs_MC.mod - 430.100000 | * observation_trends_and_prefiltering/MCMC/Trend_no_prefilter_MC.mod - 422.060000 | * observation_trends_and_prefiltering/MCMC/Trend_loglinear_no_prefilter_MC.mod - 416.320000 A full log can be found at http://www.dynare.org/testsuite/master/matlab

1 0

Status of testsuite in master branch
by Dynare Robot 15 Jun '16

15 Jun '16

b70fe6250308f8eec25f984250d75d08d74a5a01 Minor change in user reference manual. ================================ DYNARE MAKE CHECK MATLAB RESULTS ================================ | TOTAL: 690 | PASS: 680 | FAIL: 4 | XFAIL: 6 | XPASS: 0 | LIST OF FAILED TESTS: | * kalman/likelihood_from_dynare/fs2000_corr_ME.mod | * kalman/likelihood_from_dynare/fs2000_corr_ME_missing.mod | * lmmcp/sw_newton.mod | * observation_trends_and_prefiltering/ML/Trend_loglinear_prefilter_first_obs.mod | | LIST OF 10 SLOWEST TESTS: | * walsh.mod - 1152.440000 | * optimizers/fs2000_6.mod - 871.640000 | * moments/example1_hp_test.mod - 838.080000 | * ep/rbc_mc.mod - 457.910000 | * estimation/TaRB/fs2000_tarb.mod - 443.170000 | * observation_trends_and_prefiltering/MCMC/Trend_prefilter_MC.mod - 406.230000 | * observation_trends_and_prefiltering/MCMC/Trend_no_prefilter_MC.mod - 404.910000 | * observation_trends_and_prefiltering/MCMC/Trend_loglinear_no_prefilter_MC.mod - 401.530000 | * estimation/slice/fs2000_slice.mod - 397.650000 | * observation_trends_and_prefiltering/MCMC/Trend_loglinear_prefilter_MC.mod - 396.130000 A full log can be found at http://www.dynare.org/testsuite/master/matlab

1 0

fs2000_analytic_derivation.mod (testsuite)
by Stéphane Adjemian 15 Jun '16

15 Jun '16

Hi Johannes, Do we really need fs2000/fs2000_analytic_derivation.mod in the testsuite (which is failing because of non positive definite hessian matrix)? I hardly see the difference with the integration test in the analytical_derivates subfolder, except that in fs2000 subfolder we do not provide the analytical steady state. Best, Stéphane. -- Stéphane Adjemian Université du Maine, GAINS & CEPREMAP

2 2

Status of testsuite in master branch
by Dynare Robot 14 Jun '16

14 Jun '16

5814ae32e88e09d1a363eaabc9a9617767e114d6 Updated README file in tests/julia/rbc subfolder. ================================ DYNARE MAKE CHECK MATLAB RESULTS ================================ | TOTAL: 689 | PASS: 676 | FAIL: 8 | XFAIL: 5 | XPASS: 0 | LIST OF FAILED TESTS: | * identification/ident_unit_root/ident_unit_root.mod | * kalman/likelihood_from_dynare/fs2000_corr_ME.mod | * kalman/likelihood_from_dynare/fs2000_corr_ME_missing.mod | * kalman/likelihood_from_dynare/fs2000_uncorr_ME.mod | * kalman/likelihood_from_dynare/fs2000_uncorr_ME_missing.mod | * ep/rbc_mc.mod | * lmmcp/sw_newton.mod | * TeX/fs2000_corr_ME.mod | | LIST OF 10 SLOWEST TESTS: | * walsh.mod - 1154.010000 | * gsa/ls2003.mod - 960.020000 | * optimizers/fs2000_6.mod - 880.670000 | * moments/example1_hp_test.mod - 851.010000 | * estimation/TaRB/fs2000_tarb.mod - 439.960000 | * observation_trends_and_prefiltering/MCMC/Trend_no_prefilter_MC.mod - 402.340000 | * observation_trends_and_prefiltering/MCMC/Trend_loglinear_no_prefilter_MC.mod - 398.840000 | * observation_trends_and_prefiltering/MCMC/Trend_prefilter_MC.mod - 398.610000 | * estimation/slice/fs2000_slice.mod - 398.560000 | * observation_trends_and_prefiltering/MCMC/Trend_loglinear_prefilter_MC.mod - 389.650000 A full log can be found at http://www.dynare.org/testsuite/master/matlab

1 0