Dynare 4.5.7 Released
Posted on 08 February 2019We are pleased to announce the release of Dynare 4.5.7.
This is a bugfix release.
This release is compatible with MATLAB versions 7.5 (R2007b) to 9.4 (R2018b) and with GNU/Octave version 4.4.1.
Bug Fixes
Here is a list of the problems identified in version 4.5.6 and that have been fixed in version 4.5.7:

The mexfile conducting the QZ decomposition erroneously applied the
qz_criterium
to the square absolute value of eigenvalues instead of the absolute value itself (as done in mjdgges.m and the AIM solver). 
In pathological cases,
mode_compute=5
(newrat
) might enter an infinite loop. 
discretionary_policy
might erroneously state that the derivatives of the objective function are nonzero if there are NaN present. 
Dynare++, when conducting the QZ decomposition, erroneously applied the
qz_criterium
to the square absolute value of eigenvalues instead of the absolute value itself. 
Dynare++: IRFs were incorrectly computed.

dynare_sensitivity
did not display the figures ofirf_calibration
, it only stored them on the disk. 
Scatter plots generated by
dynare_sensitivity
did not correctly display LaTeX names. 
Parameter updating via steady state files did not correctly work in case of using [static]/[dynamic] equation tags.

Memory leaks in
k_order_pert
(used by higher order stochastic simulations) could lead to crashes. 
Predetermined variables were not properly set when used in model local variables.

Posterior moment computation did not correctly update the covariance matrix of exogenous shocks during posterior sampling.

Dynare was crashing with a cryptic message if a non estimated parameter was initialized in the
estimated_params_init
block. 
The
forecast
command crashed if the model was declared as linear and contained deterministic exogenous variables. 
Block decomposition is broken when used in conjunction with
varexo_det
. 
The model was not correctly specified when
identification
was run without another stochastic command in the.mod
file (e.g.estimation
,stoch_simul
, etc.). 
Realtime annualized shock decompositions added the wrong steady state value.

mh_recover
option crashed when using slice sampler. 
xaxis values in plots of moment restrictions were wrong for autocovariances.