M_.params is global and was set by csminwel1 to the last value tested; this does not necessarily coincide with the optimal parameter vector, which may have occurred earlier. This commit explicitly sets the optimal parameter vector.
Due to the preprocessor adding entries for all variable combination into i_var, variables contained in covariances obtained a wrong weighting. By only selecting the unique entries, this cannot happen anymore. Moreover, the computation intensity of the objective does not increase quadratically in the objective anymore.
- in dr_block.m and ep_residuals.m the option was hardcoded and disabled
- for csminwel1.m, no verbosity option exists anymore, thus definitions preceeding calls to it were redundant as was the options_.osr.verbose
- Adds initial check and provides error message (gets rid of cryptic error messages in OSR)
- Takes care of unaccounted errors from resol in osr_obj
- Adds function headers
by numgrad routines to compute the gradient. The default value (defined in global_initialization.m) for this parameter is 1e-6. The
value of epsilon is stored in options_.gradient_epsilon.
git-svn-id: https://www.dynare.org/svn/dynare/trunk@3335 ac1d8469-bf42-47a9-8791-bf33cf982152
preprocessor:
* add a field "M_.orig_endo_nbr" containing the nbr of endogenous before adding aux vars
* always provide "M_.aux_vars" (define it to "[]" when there is no aux var)
* rename "M_.aux_vars().orig_endo_index" to "M_.aux_vars().orig_index"
M-files:
* for commands which accept a list of variables (stoch_simul, osr, estimation, dynasave, dynatype, datatomfile), when no variable is given, use only the set of original endogenous (without aux vars) as the default
* when displaying the decision rule, when there is aux vars in the state variables, replace them by their original name (with the right lag)
* in "steady", don't display aux vars
* special exception for ramsey policy: all vars (including aux vars) are displayed, because the system of aux vars from ramsey policy is not compatible with the aux vars from the preprocessor
git-svn-id: https://www.dynare.org/svn/dynare/trunk@3166 ac1d8469-bf42-47a9-8791-bf33cf982152
* Added function numgrad3.m. This new function uses a three point formula to compute the gradient. By default Dynare uses the two point formula. By setting options_.gradient_method=3 Dynare switches to the three point formula. Note that an input argument has been added to csminwel.
* Added automatic detection of needed constant in dsge-var model (may be problematic is some cases)
git-svn-id: https://www.dynare.org/svn/dynare/dynare_v4@1935 ac1d8469-bf42-47a9-8791-bf33cf982152