2) fixed graphical and printed output for prior stab_map_.m;
3) use corrcoef for pvalue and change default pvalue_corr, alpha2_redform, alpha2_rmse to 1.e-5
* Second and third positional arguments after the name of the
estimated parameter in the estimated_params block are only
considered in the optimization stage (not in the MCMC)
* Do not store bounds in bayestopt_, because bounds do not always
reflect restrictions implied by prior shapes.
* prior_bounds routine returns a structure (with fields lb and ub)
instead of a matrix.
Sensitivity analysis for irf and moment calibration, with new function /gsa/map_calibration.m
Added new tests/gsa/ls2003a.mod for testsuite of irf/moment calibration
To be done:
- extend SA of calibration to MC and posterior samples (currently uses prior sample only)
- documentaion
- partial progress to close#267
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
1) options_gsa passed as function argument
2) use pvalues always to trigger Smirnov plots and correlation plots;
3) eliminated density plots in rmse analysis;
4) updated tex documentation
Removed globals from DsgeVarLikelihood and changed the calling sequence. As in DsgeLikelihood, the penalty is now a
persistent variable.
Added a global structure for the data: dataset_.
Removed globals from dsgevar_posterior_density and mode_check.
Simplification of the clode, definition of the variable objective_function at the top of dynare_estimation_1 (equal
to 'DsgeLikelihood' or 'DsgeVarLikelihood').