stepan
/
dynare
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge branch 'master' into ecb-master 

Fixed conflict in preprocessor/ModFile.cc
time-shift
Stéphane Adjemian (Scylla) 2017-07-27 22:48:29 +02:00
parent 7a4658323f 91096d8d02
commit c5fb6dfe57
21 changed files with 509 additions and 140 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

51
NEWS
View File

@ -1,4 +1,48 @@
Announcement for Dynare 4.5.0 (on 2013-12-16)
Announcement for Dynare 4.5.1 (on 2017-08-24)
=============================================
We are pleased to announce the release of Dynare 4.5.1.
This is a bugfix release.
The Windows packages are already available for download at:
http://www.dynare.org/download/dynare-stable
The Mac and GNU/Linux packages (for Debian and Ubuntu LTS) should follow soon.
This release is compatible with MATLAB versions 7.3 (R2006b) to 9.2 (R2017a)
and with GNU Octave versions 4.2.
Here is a list of the problems identified in version 4.5.0 and that have been
fixed in version 4.5.1:
- Fixed out of memory issue with simpsa optimization algorithm.
- Added missing plots for measurement errors with `generate_trace_plot`
command.
- Posterior moments after MCMC for very big models were not correctly computed
and their plotting might crash Dynare.
- Results of the posterior conditional variance decomposition after MCMC were
not correctly computed.
- Options `use_shock_groups` and `colormap` of the `shock_decomposition`
command were not working.
- Added a clean error message if sensitivity toolbox is used with recursive
estimation.
- Computation of posterior filtered variables was crashing in models with only
one variable.
- Fixed various typos and errors in the reference manual.
Announcement for Dynare 4.5.0 (on 2017-06-11)
=============================================
We are pleased to announce the release of Dynare 4.5.0.
@ -19,11 +63,6 @@ This release is compatible with MATLAB versions ranging from 7.3 (R2006b) to
Here is the list of major user-visible changes:
Dynare 4.5
==========
- Ramsey policy
+ Added command `ramsey_model` that builds the expanded model with

51
doc/dynare.texi
View File

@ -1220,9 +1220,10 @@ or @var{Sigma_e} to name your variable. Not conforming to this rule might yield
error messages or crashes. Second, to minimize interference with MATLAB or Octave functions that
may be called by Dynare or user-defined steady state files, it is recommended to avoid using the name
of MATLAB functions. In particular when working with steady state files, do not use correctly-spelled greek
names like @var{alpha}, because there are Matlab functions of the same name. Rather go for @var{alppha} or so.
names like @var{alpha}, because there are Matlab functions of the same name. Rather go for @var{alppha} or @var{alph}.
Lastly, please do not name a variable or parameter @var{i}. This may interfere with the imaginary
number @var{i} and the index in many loops. Rather, name investment @var{invest}.
number @var{i} and the index in many loops. Rather, name investment @var{invest}. Using @var{inv} is also not recommended
as it alread denotes the inverse operator.
Declarations of variables and parameters are made with the following commands:
@ -1984,7 +1985,7 @@ different purposes by allowing the user to attach arbitrary informations to each
equation and to recover them at runtime. For instance, it is possible to name the
equations with a @code{name}-tag, using a syntax like:
@example
mode;
model;
...
[name = 'Budget constraint']
c + k = k^theta*A;
@ -1996,7 +1997,7 @@ of the model is tagged with a name, the @code{resid} command
will display the name of the equations (which may be more informative than the
equation numbers) in addition to the equation number. Several tags for one equation can be separated using a comma.
@example
mode;
model;
...
[name='Taylor rule',mcp = 'r > -1.94478']
r = rho*r(-1) + (1-rho)*(gpi*Infl+gy*YGap) + e;
@ -2362,7 +2363,7 @@ static equilibria, but not necessarily.
One typical application is to consider an economy at the equilibrium at time 0,
trigger a shock in first period, and study the trajectory of return to
the initial equilibrium. To do that, one needs @code{initval} and
@code{shocks} (@pxref{Shocks on exogenous variables}.
@code{shocks} (@pxref{Shocks on exogenous variables}).
Another one is to study, how an economy, starting from arbitrary
initial conditions at time 0 converges toward equilibrium.
@ -3233,7 +3234,7 @@ option, @pxref{Model declaration})
Trust-region algorithm on the entire model.
@item 10
Levenberg-Marquardt mixed compleproblem (LMMCP) solver
Levenberg-Marquardt mixed complementarity problem (LMMCP) solver
(@cite{Kanzow and Petra 2004})
@item 11
@ -3680,7 +3681,7 @@ to the equation appears currently on the left hand side and where no
backward looking endogenous variables appear. The block has the form:
@math{y_{j,t} = f_j(y_t, y_{t+1}, \ldots, y_{t+k})}.
@item SOLVE FORWARD @var{x}
@item SOLVE BACKWARD @var{x}
The block contains only equations where endogenous variable attributed
to the equation does not appear currently on the left hand side and
where no forward looking endogenous variables appear. The block has
@ -4386,7 +4387,7 @@ If option @code{periods} is present, sets @code{oo_.skewness},
MATLAB/Octave vectors of the global workspace with the same name as
the endogenous variables.
If options @code{irf} is different from zero, sets @code{oo_.irfs}
If option @code{irf} is different from zero, sets @code{oo_.irfs}
(see below) and also saves the IRFs in MATLAB/Octave vectors of
the global workspace (this latter way of accessing the IRFs is
deprecated and will disappear in a future version).
@ -4812,7 +4813,7 @@ non-decreasing tuples are stored, @i{i.e.} those for which @math{i_1
\leq i_2 \leq i_3}. The columns are arranged in the lexicographical
order of non-decreasing tuples. Also note that for tuples that have
three distinct indices (@i{i.e.} @math{i_1 \neq i_2} and @math{i_1
\neq i_3} and @math{i_2 \neq i_3}, since these elements are stored
\neq i_3} and @math{i_2 \neq i_3}), since these elements are stored
only once but appears six times in the unfolded @math{G_3} matrix,
they must be multiplied by 6 when computing the decision
rules. Similarly, for those tuples that have two equal indices
@ -4992,7 +4993,7 @@ likelihood estimation. In a Bayesian estimation context, sets a lower bound
only effective while maximizing the posterior kernel. This lower bound does not
modify the shape of the prior density, and is only aimed at helping the
optimizer in identifying the posterior mode (no consequences for the MCMC). For
some prior densities (namely inverse gamma, gamma, uniform, beta or weibull) it is
some prior densities (namely inverse gamma, gamma, uniform, beta or Weibull) it is
possible to shift the support of the prior distributions to the left or the right using
@ref{prior_3rd_parameter}. In this case the prior density is effectively
modified (note that the truncated Gaussian density is not implemented in
@ -5020,7 +5021,7 @@ that @code{inv_gamma_pdf} is equivalent to
@item @var{PRIOR_3RD_PARAMETER}
@anchor{prior_3rd_parameter}
A third parameter of the prior used for generalized beta distribution,
generalized gamma, generalized weibull and for the uniform distribution. Default: @code{0}
generalized gamma, generalized Weibull and for the uniform distribution. Default: @code{0}
@item @var{PRIOR_4TH_PARAMETER}
@anchor{prior_4th_parameter}
@ -5921,7 +5922,7 @@ distribution of IRFs. The length of the IRFs are controlled by the
weight of the DSGE prior of the VAR model is calibrated to the value
passed (see @cite{Del Negro and Schorfheide (2004)}). It represents ratio of dummy over actual observations.
To assure that the prior is proper, the value must be bigger than @math{(k+n)/T},
where @math{k} is the number of estimated parameters, @math{n} is the number of observables, #
where @math{k} is the number of estimated parameters, @math{n} is the number of observables,
and @math{T} is the number of observations. NB: The previous method
of declaring @code{dsge_prior_weight} as a parameter and then
calibrating it is now deprecated and will be removed in a future release
@ -6833,8 +6834,8 @@ observation for which the forecast has been made.
@defvr {MATLAB/Octave variable} oo_.Smoother.State_uncertainty
@anchor{oo_.Smoother.State_uncertainty}
Three-dimensional array set by the @code{estimation} command (if used with the
@code{smoother}) without Metropolis,
or by the @code{calib_smoother} command, if the @code{o_smoothed_state_uncertainty} option
@code{smoother} option) without Metropolis,
or by the @code{calib_smoother} command, if the @code{smoothed_state_uncertainty} option
has been requested.
Contains the series of covariance matrices for the state estimate given the full data
from the Kalman smoother. The @code{M_.endo_nbr} times @code{M_.endo_nbr} times
@ -7775,7 +7776,7 @@ uncertainty, but ignoring the effect of measurement error on
observed variables
@item HPDsup
Lower bound of a 90% HPD interval due to parameter uncertainty, but
Upper bound of a 90% HPD forecast interval due to parameter uncertainty, but
ignoring the effect of measurement error on
observed variables
@ -7785,7 +7786,7 @@ to change the size of the HPD interval} of forecast for observed variables
due to parameter uncertainty and measurement error
@item HPDsup_ME
Lower bound of a 90% HPD interval of forecast for observed variables
Upper bound of a 90% HPD interval of forecast for observed variables
due to parameter uncertainty and measurement error
@item Mean
@ -8385,7 +8386,7 @@ Note that the use of this block requires the use of a constrained optimizer, @i{
@example
osr_param_bounds;
osr_params_bounds;
gamma_inf_, 0, 2.5;
end;
@ -9275,7 +9276,7 @@ named as:
@code{<namendo>_vs_<namexo>}: for entries of the matrix of the shocks.
@end itemize
The following files are stored in each directory (we stick with prior sample but similar convetins are used for MC samples):
The following files are stored in each directory (we stick with prior sample but similar conventions are used for MC samples):
@itemize
@item
@code{<mod_file>_prior_<namendo>_vs_<namexo>.fig}: histogram and CDF plot of the MC sample of the individual entry
@ -9551,7 +9552,7 @@ In case of estimating a MS-DSGE model,@footnote{An example can be found at @uref
@table @code
@item parameters = @var{[LIST OF PARAMETERS]}
This option specifies which parameters are controlled by this Markov Chai
This option specifies which parameters are controlled by this Markov Chain.
@item number_of_lags = @var{DOUBLE}
Provides the number of lags that each parameter can take within each regime in this chain.
@ -10970,8 +10971,8 @@ This file computes the steady state. It begins with:
@@#include "modeqs.mod"
@end example
Then it initializes parameters (including @code{lab_rat}, excluding
@math{\alpha}, computes the steady state (using guess values for
endogenous, including @math{\alpha}, then saves values of parameters
@math{\alpha}), computes the steady state (using guess values for
endogenous, including @math{\alpha}), then saves values of parameters
and endogenous at steady state in a file, using the
@code{save_params_and_steady_state} command.
@ -11028,7 +11029,7 @@ array (@code{rhos}).
The advantage of this method is that it uses a shorter syntax, since
list of values directly given in the loop construct. Note that values
are given as character strings (the macro-processor does not know
floating point values. The inconvenient is that you can not reuse an
floating point values). The inconvenience is that you can not reuse an
array stored in a MATLAB/Octave variable.
@end table
@ -11987,7 +11988,7 @@ ans =
@deftypefn{dates} {@var{C} =} gt (@var{A}, @var{B})
Overloads the Matlab/Octave @code{gt} (greater than, @code{>=}) operator. @dates objects @var{A} and @var{B} must have the same number of elements (say, @code{n}). The returned argument is a @code{n} by @code{1} vector of zeros and ones. The i-th element of @var{C} is equal to @code{1} if and only if the date @code{A(i)} is posterior to the date @code{B(i)}.
Overloads the Matlab/Octave @code{gt} (greater than, @code{>}) operator. @dates objects @var{A} and @var{B} must have the same number of elements (say, @code{n}). The returned argument is a @code{n} by @code{1} vector of zeros and ones. The i-th element of @var{C} is equal to @code{1} if and only if the date @code{A(i)} is posterior to the date @code{B(i)}.
@examplehead
@example
@ -12136,7 +12137,7 @@ ans =
@deftypefn{dates} {@var{C} =} lt (@var{A}, @var{B})
Overloads the Matlab/Octave @code{lt} (less than, @code{<=}) operator. @dates objects @var{A} and @var{B} must have the same number of elements (say, @code{n}). The returned argument is a @code{n} by @code{1} vector of zeros and ones. The i-th element of @var{C} is equal to @code{1} if and only if the date @code{A(i)} preceeds the date @code{B(i)}.
Overloads the Matlab/Octave @code{lt} (less than, @code{<}) operator. @dates objects @var{A} and @var{B} must have the same number of elements (say, @code{n}). The returned argument is a @code{n} by @code{1} vector of zeros and ones. The i-th element of @var{C} is equal to @code{1} if and only if the date @code{A(i)} preceeds the date @code{B(i)}.
@examplehead
@example
@ -14010,7 +14011,7 @@ command. Default: @code{`!'}
@end defmethod
@anchor{addGraph}
@defmethod Report addGraph axisShape, data, graphDirName, graphName, graphSize, height, showGrid, showLegend, legendAt, showLegendBox, legendLocation, legendOrientation, legendFontSize, miscTikzAxisOptions, miscTikzPictureOptions, seriesToUse, shade, shadeColor, shadeOpacity, tickFontSize, title, titleFontSize, titleFormat, width, writeCSV, xlabel, ylabel, xAxisTight, xrange, xTicks, xTickLabels, xTickLabelAnchor, xTickLabelRotation, yAxisTight, yTickLabelFixed, yTickLabelPrecision, yTickLabelScaled, yTickLabelZeroFill, yrange, showZeroline, zeroLineColor
@defmethod Report addGraph axisShape, data, graphDirName, graphName, graphSize, height, showGrid, showLegend, legendAt, showLegendBox, legendLocation, legendOrientation, legendFontSize, miscTikzAxisOptions, miscTikzPictureOptions, seriesToUse, shade, shadeColor, shadeOpacity, tickFontSize, title, titleFontSize, titleFormat, width, writeCSV, xlabel, ylabel, xAxisTight, xrange, xTicks, xTickLabels, xTickLabelAnchor, xTickLabelRotation, yAxisTight, yTickLabelFixed, yTickLabelPrecision, yTickLabelScaled, yTickLabelZeroFill, yrange, showZeroLine, zeroLineColor
Adds a @code{Graph} to a @code{Section}.
@optionshead
@table @code

9
matlab/discretionary_policy_1.m
View File

@ -68,6 +68,15 @@ if options_.steadystate_flag
end
[U,Uy,W] = feval([M_.fname,'_objective_static'],zeros(endo_nbr,1),[], M_.params);
if any(any(Uy~=0))
non_zero_derivs=find(any(Uy~=0));
for ii=1:length(non_zero_derivs)
non_zero_deriv_names{ii,1}=deblank(M_.endo_names(non_zero_derivs(ii),:));
end
disp_string=[non_zero_deriv_names{1,:}];
for ii=2:size(non_zero_deriv_names,1)
disp_string=[disp_string,', ',non_zero_deriv_names{ii,:}];
end
fprintf('\nThe derivative of the objective function w.r.t. to variable(s) %s is not 0\n',disp_string)
error(['discretionary_policy: the objective function must have zero ' ...
'first order derivatives'])
end

4
matlab/dsge_simulated_theoretical_conditional_variance_decomposition.m
View File

@ -109,10 +109,10 @@ for file = 1:NumberOfDrawsFiles
for linee = 1:NumberOfDraws
linea = linea+1;
if isdrsaved
set_parameters(pdraws{linee,1});% Needed to update the covariance matrix of the state innovations.
M_=set_parameters_locally(M_,pdraws{linee,1});% Needed to update the covariance matrix of the state innovations.
dr = pdraws{linee,2};
else
set_parameters(pdraws{linee,1});
M_=set_parameters_locally(M_,pdraws{linee,1});
[dr,info,M_,options_,oo_] = resol(0,M_,options_,oo_);
end
if first_call

2
matlab/dsge_simulated_theoretical_correlation.m
View File

@ -106,7 +106,7 @@ for file = 1:NumberOfDrawsFiles
if isdrsaved
dr = pdraws{linee,2};
else
set_parameters(pdraws{linee,1});
M_=set_parameters_locally(M_,pdraws{linee,1});
[dr,info,M_,options_,oo_] = resol(0,M_,options_,oo_);
end
tmp = th_autocovariances(dr,ivar,M_,options_,nodecomposition);

2
matlab/dsge_simulated_theoretical_covariance.m
View File

@ -105,7 +105,7 @@ for file = 1:NumberOfDrawsFiles
if isdrsaved
dr = pdraws{linee,2};
else
set_parameters(pdraws{linee,1});
M_=set_parameters_locally(M_,pdraws{linee,1});
[dr,info,M_,options_,oo_] = resol(0,M_,options_,oo_);
end
tmp = th_autocovariances(dr,ivar,M_,options_,nodecomposition);

2
matlab/dsge_simulated_theoretical_variance_decomposition.m
View File

@ -113,7 +113,7 @@ for file = 1:NumberOfDrawsFiles
if isdrsaved
dr = pdraws{linee,2};
else
set_parameters(pdraws{linee,1});
M_=set_parameters_locally(M_,pdraws{linee,1});
[dr,info,M_,options_,oo_] = resol(0,M_,options_,oo_);
end
if file==1 && linee==1

5
matlab/generate_trace_plots.m
View File

@ -50,6 +50,11 @@ for ii=1:size(estim_params_.var_exo,1)
trace_plot(options_,M_,estim_params_,'StructuralShock',chain_number,parameter_name)
end
for ii=1:size(estim_params_.var_endo,1)
parameter_name=deblank(M_.endo_names(estim_params_.var_endo(ii,1),:));
trace_plot(options_,M_,estim_params_,'MeasurementError',chain_number,parameter_name)
end
for ii=1:size(estim_params_.corrn,1)
parameter_name_1=deblank(M_.endo_names(estim_params_.corrn(ii,1),:));
parameter_name_2=deblank(M_.endo_names(estim_params_.corrn(ii,2),:));

71
matlab/optimization/simpsa.m
View File

@ -142,8 +142,9 @@ DEFAULT_OPTIONS = simpsaset('TEMP_START',[],... % starting temperature (if none
OPTIONS = simpsaset(DEFAULT_OPTIONS,OPTIONS);
% store options in OUTPUT
OUTPUT.OPTIONS = OPTIONS;
if nargout>3
OUTPUT.OPTIONS = OPTIONS;
end
% initialize simplex
% ------------------
@ -175,13 +176,13 @@ end
% initialize OUTPUT structure
% ---------------------------
OUTPUT.TEMPERATURE = zeros(OPTIONS.MAX_ITER_TOTAL,1);
OUTPUT.SIMPLEX = zeros(NDIM+1,NDIM,OPTIONS.MAX_ITER_TOTAL);
OUTPUT.SIMPLEX_BEST = zeros(OPTIONS.MAX_ITER_TOTAL,NDIM);
OUTPUT.COSTS = zeros(OPTIONS.MAX_ITER_TOTAL,NDIM+1);
OUTPUT.COST_BEST = zeros(OPTIONS.MAX_ITER_TOTAL,1);
if nargout>3
OUTPUT.TEMPERATURE = zeros(OPTIONS.MAX_ITER_TOTAL,1);
OUTPUT.SIMPLEX = zeros(NDIM+1,NDIM,OPTIONS.MAX_ITER_TOTAL);
OUTPUT.SIMPLEX_BEST = zeros(OPTIONS.MAX_ITER_TOTAL,NDIM);
OUTPUT.COSTS = zeros(OPTIONS.MAX_ITER_TOTAL,NDIM+1);
OUTPUT.COST_BEST = zeros(OPTIONS.MAX_ITER_TOTAL,1);
end
% initialize iteration data
% -------------------------
@ -304,17 +305,19 @@ while 1
Y = help(:,2);
P = help(:,3:end);
% store temperature at current iteration
OUTPUT.TEMPERATURE(nITERATIONS) = TEMP;
if nargout>3
% store temperature at current iteration
OUTPUT.TEMPERATURE(nITERATIONS) = TEMP;
% store information about simplex at the current iteration
OUTPUT.SIMPLEX(:,:,nITERATIONS) = P;
OUTPUT.SIMPLEX_BEST(nITERATIONS,:) = PBEST;
% store cost function value of best vertex in current iteration
OUTPUT.COSTS(nITERATIONS,:) = Y;
OUTPUT.COST_BEST(nITERATIONS) = YBEST;
% store information about simplex at the current iteration
OUTPUT.SIMPLEX(:,:,nITERATIONS) = P;
OUTPUT.SIMPLEX_BEST(nITERATIONS,:) = PBEST;
% store cost function value of best vertex in current iteration
OUTPUT.COSTS(nITERATIONS,:) = Y;
OUTPUT.COST_BEST(nITERATIONS) = YBEST;
end
if strcmp(OPTIONS.DISPLAY,'iter')
disp(sprintf('%5.0f %5.0f %12.6g %15.6g %12.6g %s',nITERATIONS,nFUN_EVALS,Y(1),YBEST,TEMP,ALGOSTEP));
end
@ -452,21 +455,23 @@ end
X = transpose(PBEST);
FVAL = YBEST;
% store number of function evaluations
OUTPUT.nFUN_EVALS = nFUN_EVALS;
% store number of iterations
OUTPUT.nITERATIONS = nITERATIONS;
% trim OUTPUT data structure
OUTPUT.TEMPERATURE(nITERATIONS+1:end) = [];
OUTPUT.SIMPLEX(:,:,nITERATIONS+1:end) = [];
OUTPUT.SIMPLEX_BEST(nITERATIONS+1:end,:) = [];
OUTPUT.COSTS(nITERATIONS+1:end,:) = [];
OUTPUT.COST_BEST(nITERATIONS+1:end) = [];
% store the amount of time needed in OUTPUT data structure
OUTPUT.TIME = toc;
if nargout>3
% store number of function evaluations
OUTPUT.nFUN_EVALS = nFUN_EVALS;
% store number of iterations
OUTPUT.nITERATIONS = nITERATIONS;
% trim OUTPUT data structure
OUTPUT.TEMPERATURE(nITERATIONS+1:end) = [];
OUTPUT.SIMPLEX(:,:,nITERATIONS+1:end) = [];
OUTPUT.SIMPLEX_BEST(nITERATIONS+1:end,:) = [];
OUTPUT.COSTS(nITERATIONS+1:end,:) = [];
OUTPUT.COST_BEST(nITERATIONS+1:end) = [];
% store the amount of time needed in OUTPUT data structure
OUTPUT.TIME = toc;
end
return

2
matlab/osr1.m
View File

@ -126,7 +126,7 @@ elseif isequal(options_.osr.opt_algo,11)
else
if ~isempty(M_.osr.param_bounds) && ~(ismember(options_.osr.opt_algo,[1,2,5,9]) || ischar(options_.osr.opt_algo))
error('OSR: OSR with bounds on parameters requires a constrained optimizer, i.e. 1,2,5, or 9.')
error('OSR: OSR with bounds on parameters requires a constrained optimizer, i.e. opt_algo= 1,2,5, or 9.')
end
%%do actual optimization
[p, f, exitflag] = dynare_minimize_objective(str2func('osr_obj'),t0,options_.osr.opt_algo,options_,M_.osr.param_bounds,cellstr(M_.param_names(i_params,:)),[],[], i_params,...

1
matlab/plot_shock_decomposition.m
View File

@ -30,7 +30,6 @@ function [z, steady_state] = plot_shock_decomposition(M_,oo_,options_,varlist)
options_.nodisplay = options_.plot_shock_decomp.nodisplay;
options_.graph_format = options_.plot_shock_decomp.graph_format;
options_.use_shock_groups = options_.plot_shock_decomp.use_shock_groups;
% indices of endogenous variables
if size(varlist,1) == 0

16
matlab/prior_posterior_statistics_core.m
View File

@ -260,23 +260,23 @@ for b=fpar:B
else
constant_part=repmat(SteadyState(dr.order_var)',[length(options_.filter_step_ahead),1,gend+max(options_.filter_step_ahead)]);
end
stock_filter_step_ahead(:,dr.order_var,:,irun(4)) = aK(options_.filter_step_ahead,1:endo_nbr,:) + ...
constant_part;
stock_filter_step_ahead(:,dr.order_var,:,irun(4)) = aK(options_.filter_step_ahead,1:endo_nbr,:) + constant_part;
%now add trend to observables
for ii=1:length(options_.filter_step_ahead)
if options_.prefilter
stock_filter_step_ahead(ii,IdObs,:,irun(4)) = squeeze(stock_filter_step_ahead(ii,IdObs,:,irun(4)))...
+repmat(mean_correction(:,1),1,gend+max(options_.filter_step_ahead))... %constant correction
zdim = size(stock_filter_step_ahead(ii,IdObs,:,irun(4)));
squeezed = reshape(stock_filter_step_ahead(ii,IdObs,:,irun(4)), [zdim(2:end) 1]);
stock_filter_step_ahead(ii,IdObs,:,irun(4)) = squeezed ...
+repmat(mean_correction(:,1),1,gend+max(options_.filter_step_ahead)) ... %constant correction
+[trend_addition repmat(trend_addition(:,end),1,max(options_.filter_step_ahead))+trend_coeff*[1:max(options_.filter_step_ahead)]]; %trend
else
stock_filter_step_ahead(ii,IdObs,:,irun(4)) = squeeze(stock_filter_step_ahead(ii,IdObs,:,irun(4)))...
zdim = size(stock_filter_step_ahead(ii,IdObs,:,irun(4)));
squeezed = reshape(stock_filter_step_ahead(ii,IdObs,:,irun(4)), [zdim(2:end) 1]);
stock_filter_step_ahead(ii,IdObs,:,irun(4)) = squeezed ...
+[trend_addition repmat(trend_addition(:,end),1,max(options_.filter_step_ahead))+trend_coeff*[1:max(options_.filter_step_ahead)]]; %trend
end
end
end
if horizon
yyyy = alphahat(iendo,i_last_obs);
yf = forcst2a(yyyy,dr,zeros(horizon,exo_nbr));

51
matlab/set_parameters.m
View File

@ -31,53 +31,6 @@ function set_parameters(xparam1)
% You should have received a copy of the GNU General Public License
% along with Dynare. If not, see <http://www.gnu.org/licenses/>.
global estim_params_ M_
global M_
nvx = estim_params_.nvx;
ncx = estim_params_.ncx;
nvn = estim_params_.nvn;
ncn = estim_params_.ncn;
np = estim_params_.np;
Sigma_e = M_.Sigma_e;
Correlation_matrix = M_.Correlation_matrix;
offset = 0;
% setting shocks variance on the diagonal of Covariance matrix; used later
% for updating covariances
if nvx
var_exo = estim_params_.var_exo;
for i=1:nvx
k = var_exo(i,1);
Sigma_e(k,k) = xparam1(i)^2;
end
end
% and update offset
offset = offset + nvx + nvn;
% correlations amonx shocks (ncx)
if ncx
corrx = estim_params_.corrx;
for i=1:ncx
k1 = corrx(i,1);
k2 = corrx(i,2);
Correlation_matrix(k1,k2) = xparam1(i+offset);
Correlation_matrix(k2,k1) = Correlation_matrix(k1,k2);
end
end
%build covariance matrix from correlation matrix and variances already on
%diagonal
Sigma_e = diag(sqrt(diag(Sigma_e)))*Correlation_matrix*diag(sqrt(diag(Sigma_e)));
if isfield(estim_params_,'calibrated_covariances')
Sigma_e(estim_params_.calibrated_covariances.position)=estim_params_.calibrated_covariances.cov_value;
end
% and update offset
offset = offset + ncx + ncn;
% structural parameters
if np
M_.params(estim_params_.param_vals(:,1)) = xparam1(offset+1:end);
end
M_.Sigma_e = Sigma_e;
M_.Correlation_matrix=Correlation_matrix;
M_ = set_parameters_locally(M_, xparam1);

85
matlab/set_parameters_locally.m Normal file
View File

@ -0,0 +1,85 @@
function M_=set_parameters_locally(M_,xparam1)
% function M_out=set_parameters(M_,xparam1)
% Sets parameters value (except measurement errors)
% This is called for computations such as IRF and forecast
% when measurement errors aren't taken into account; in contrast to
% set_parameters.m, the global M_-structure is not altered
%
% INPUTS
% xparam1: vector of parameters to be estimated (initial values)
% M_: Dynare model-structure
%
% OUTPUTS
% M_: Dynare model-structure
%
% SPECIAL REQUIREMENTS
% none
% Copyright (C) 2017 Dynare Team
%
% This file is part of Dynare.
%
% Dynare is free software: you can redistribute it and/or modify
% it under the terms of the GNU General Public License as published by
% the Free Software Foundation, either version 3 of the License, or
% (at your option) any later version.
%
% Dynare is distributed in the hope that it will be useful,
% but WITHOUT ANY WARRANTY; without even the implied warranty of
% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
% GNU General Public License for more details.
%
% You should have received a copy of the GNU General Public License
% along with Dynare. If not, see <http://www.gnu.org/licenses/>.
global estim_params_
nvx = estim_params_.nvx;
ncx = estim_params_.ncx;
nvn = estim_params_.nvn;
ncn = estim_params_.ncn;
np = estim_params_.np;
Sigma_e = M_.Sigma_e;
Correlation_matrix = M_.Correlation_matrix;
offset = 0;
% setting shocks variance on the diagonal of Covariance matrix; used later
% for updating covariances
if nvx
var_exo = estim_params_.var_exo;
for i=1:nvx
k = var_exo(i,1);
Sigma_e(k,k) = xparam1(i)^2;
end
end
% and update offset
offset = offset + nvx + nvn;
% correlations amonx shocks (ncx)
if ncx
corrx = estim_params_.corrx;
for i=1:ncx
k1 = corrx(i,1);
k2 = corrx(i,2);
Correlation_matrix(k1,k2) = xparam1(i+offset);
Correlation_matrix(k2,k1) = Correlation_matrix(k1,k2);
end
end
%build covariance matrix from correlation matrix and variances already on
%diagonal
Sigma_e = diag(sqrt(diag(Sigma_e)))*Correlation_matrix*diag(sqrt(diag(Sigma_e)));
if isfield(estim_params_,'calibrated_covariances')
Sigma_e(estim_params_.calibrated_covariances.position)=estim_params_.calibrated_covariances.cov_value;
end
% and update offset
offset = offset + ncx + ncn;
% structural parameters
if np
M_.params(estim_params_.param_vals(:,1)) = xparam1(offset+1:end);
end
M_.Sigma_e = Sigma_e;
M_.Correlation_matrix=Correlation_matrix;

65
preprocessor/DynamicModel.cc
View File

@ -44,6 +44,10 @@ DynamicModel::DynamicModel(SymbolTable &symbol_table_arg,
max_endo_lag(0), max_endo_lead(0),
max_exo_lag(0), max_exo_lead(0),
max_exo_det_lag(0), max_exo_det_lead(0),
max_lag_orig(0), max_lead_orig(0),
max_endo_lag_orig(0), max_endo_lead_orig(0),
max_exo_lag_orig(0), max_exo_lead_orig(0),
max_exo_det_lag_orig(0), max_exo_det_lead_orig(0),
dynJacobianColsNbr(0),
global_temporary_terms(true)
{
@ -2594,7 +2598,15 @@ DynamicModel::writeOutput(ostream &output, const string &basename, bool block_de
outstruct = "oo_.";
}
output << modstruct << "lead_lag_incidence = [";
output << modstruct << "max_endo_lag_orig = " << max_endo_lag_orig << ";" << endl
<< modstruct << "max_endo_lead_orig = " << max_endo_lead_orig << ";" << endl
<< modstruct << "max_exo_lag_orig = " << max_exo_lag_orig << ";" << endl
<< modstruct << "max_exo_lead_orig = " << max_exo_lead_orig << ";" << endl
<< modstruct << "max_exo_det_lag_orig = " << max_exo_det_lag_orig << ";" << endl
<< modstruct << "max_exo_det_lead_orig = " << max_exo_det_lead_orig << ";" << endl
<< modstruct << "max_lag_orig = " << max_lag_orig << ";" << endl
<< modstruct << "max_lead_orig = " << max_lead_orig << ";" << endl
<< modstruct << "lead_lag_incidence = [";
// Loop on endogenous variables
int nstatic = 0,
nfwrd = 0,
@ -3830,6 +3842,8 @@ DynamicModel::cloneDynamic(DynamicModel &dynamic_model) const
for (size_t i = 0; i < static_only_equations.size(); i++)
dynamic_model.addStaticOnlyEquation(static_only_equations[i]->cloneDynamic(dynamic_model),
static_only_equations_lineno[i]);
dynamic_model.setLeadsLagsOrig();
}
void
@ -4008,6 +4022,55 @@ DynamicModel::findUnusedExogenous()
return unusedExo;
}
void
DynamicModel::setLeadsLagsOrig()
{
set<pair<int, int> > dynvars;
for (int i = 0; i < (int) equations.size(); i++)
{
equations[i]->collectDynamicVariables(eEndogenous, dynvars);
equations[i]->collectDynamicVariables(eExogenous, dynvars);
equations[i]->collectDynamicVariables(eExogenousDet, dynvars);
}
for (set<pair<int, int> >::const_iterator it = dynvars.begin();
it != dynvars.end(); it++)
{
int lag = it->second;
SymbolType type = symbol_table.getType(it->first);
if (max_lead_orig < lag)
max_lead_orig= lag;
else if (-max_lag_orig > lag)
max_lag_orig = -lag;
switch (type)
{
case eEndogenous:
if (max_endo_lead_orig < lag)
max_endo_lead_orig = lag;
else if (-max_endo_lag_orig > lag)
max_endo_lag_orig = -lag;
break;
case eExogenous:
if (max_exo_lead_orig < lag)
max_exo_lead_orig = lag;
else if (-max_exo_lag_orig > lag)
max_exo_lag_orig = -lag;
break;
case eExogenousDet:
if (max_exo_det_lead_orig < lag)
max_exo_det_lead_orig = lag;
else if (-max_exo_det_lag_orig > lag)
max_exo_det_lag_orig = -lag;
break;
default:
break;
}
}
}
void
DynamicModel::computeDerivIDs()
{

11
preprocessor/DynamicModel.hh
View File

@ -61,6 +61,14 @@ private:
//! Maximum lag and lead over deterministic exogenous variables (positive values)
/*! Set by computeDerivIDs() */
int max_exo_det_lag, max_exo_det_lead;
//! Maximum lag and lead over all types of variables (positive values) of original model
int max_lag_orig, max_lead_orig;
//! Maximum lag and lead over endogenous variables (positive values) of original model
int max_endo_lag_orig, max_endo_lead_orig;
//! Maximum lag and lead over exogenous variables (positive values) of original model
int max_exo_lag_orig, max_exo_lead_orig;
//! Maximum lag and lead over deterministic exogenous variables (positive values) of original model
int max_exo_det_lag_orig, max_exo_det_lead_orig;
//! Cross reference information
map<int, ExprNode::EquationInfo> xrefs;
@ -287,6 +295,9 @@ public:
//! Find exogenous variables not used in model
set<int> findUnusedExogenous();
//! Set the max leads/lags of the original model
void setLeadsLagsOrig();
//! Copies a dynamic model (only the equations)
/*! It assumes that the dynamic model given in argument has just been allocated */
void cloneDynamic(DynamicModel &dynamic_model) const;

12
preprocessor/DynareBison.yy
View File

@ -2688,8 +2688,8 @@ plot_shock_decomposition_options_list : plot_shock_decomposition_option COMMA pl
| plot_shock_decomposition_option
;
plot_shock_decomposition_option : o_psd_use_shock_groups
| o_psd_colormap
plot_shock_decomposition_option : o_use_shock_groups
| o_colormap
| o_psd_nodisplay
| o_psd_graph_format
| o_psd_detail_plot
@ -3514,14 +3514,10 @@ o_lmmcp : LMMCP {driver.option_num("lmmcp.status", "1"); };
o_occbin : OCCBIN {driver.option_num("occbin", "1"); };
o_function : FUNCTION EQUAL filename { driver.option_str("function", $3); };
o_sampling_draws : SAMPLING_DRAWS EQUAL INT_NUMBER { driver.option_num("sampling_draws",$3); };
o_use_shock_groups : USE_SHOCK_GROUPS { driver.option_str("use_shock_groups","default"); }
| USE_SHOCK_GROUPS EQUAL symbol { driver.option_str("use_shock_groups", $3); }
o_use_shock_groups : USE_SHOCK_GROUPS { driver.option_str("plot_shock_decomp.use_shock_groups","default"); }
| USE_SHOCK_GROUPS EQUAL symbol { driver.option_str("plot_shock_decomp.use_shock_groups", $3); }
;
o_psd_use_shock_groups : USE_SHOCK_GROUPS { driver.option_str("plot_shock_decomp.use_shock_groups","default"); }
| USE_SHOCK_GROUPS EQUAL symbol { driver.option_str("plot_shock_decomp.use_shock_groups", $3); }
;
o_colormap : COLORMAP EQUAL symbol { driver.option_num("plot_shock_decomp.colormap",$3); };
o_psd_colormap : COLORMAP EQUAL symbol { driver.option_num("plot_shock_decomp.colormap",$3); };
range : symbol ':' symbol
{

9
scripts/dynare.el
View File

@ -28,8 +28,6 @@
;; add this to your .emacs or site-init.el file:
;;
;; (require 'dynare)
;; (autoload 'dynare-mode "dynare" "Enter dynare mode." t)
;; (setq auto-mode-alist (cons '("\\.mod\\'" . dynare-mode) auto-mode-alist))
;;; Commentary:
;;
@ -107,7 +105,8 @@ For detail, see `comment-dwim'."
("(\\(+\\|-\\)[1-9])" . font-lock-constant-face)
))
;; define the major mode
;;; define the major mode
;;;###autoload
(define-derived-mode dynare-mode fundamental-mode
"dynare mode"
"dynare is a mode for editing mod files used by dynare."
@ -149,5 +148,9 @@ For detail, see `comment-dwim'."
(setq dynare-functions-regexp nil)
)
;;; mode trigger
;;;###autoload
(add-to-list 'auto-mode-alist '("\\.mod$" . dynare-mode))
(provide 'dynare)
;;; dynare.el ends here

2
tests/Makefile.am
View File

@ -302,6 +302,7 @@ MODFILES = \
gradient/fs2000_numgrad_5.mod \
filter_step_ahead/fs2000_filter_step_ahead_bayesian.mod \
filter_step_ahead/fs2000_filter_step_ahead_ML.mod \
filter_step_ahead/trend_cycle_decomposition.mod \
loglinear/example4_exp.mod \
loglinear/example4_loglinear.mod \
loglinear/example4_loglinear_lagged_exogenous.mod \
@ -762,6 +763,7 @@ EXTRA_DIST = \
external_function/extFunNoDerivs.m \
external_function/extFunWithFirstAndSecondDerivs.m \
expectations/expectation_ss_old_steadystate.m \
filter_step_ahead/trend_cycle_decomposition_data.m \
steady_state/walsh1_old_ss_steadystate.m \
data/test.xlsx \
gsa/morris/nk_est_data.m \

34
tests/filter_step_ahead/trend_cycle_decomposition.mod Normal file
View File

@ -0,0 +1,34 @@
var y yp z mu;
varexo ez eyp emu;
parameters alpha;
alpha = .889;
model(linear);
y = yp + z;
yp = mu + yp(-1) + eyp;
mu = mu(-1) + emu;
z = alpha*z(-1) + ez;
end;
initval;
y=8.655680;
z=0;
yp=8.655680;
mu=0;
end;
steady(nocheck);
estimated_params;
stderr emu , inv_gamma_pdf, 0.002 , inf;
stderr eyp , inv_gamma_pdf, 0.002 , inf;
stderr ez , inv_gamma_pdf, 0.06 , inf;
alpha, normal_pdf, 0.9, 0.1;
end;
varobs y;
estimation(datafile=trend_cycle_decomposition_data,nobs=82, mh_replic=2000, mode_compute=4, mh_nblocks=1, mh_jscale=0.3, filtered_vars, smoother, diffuse_filter) yp z;

164
tests/filter_step_ahead/trend_cycle_decomposition_data.m Normal file
View File

@ -0,0 +1,164 @@
y=[8.655680
8.656123
8.654401
8.641229
8.630567
8.620637
8.605288
8.608660
8.620357
8.619683
8.630202
8.636380
8.614226
8.605106
8.537263
8.555079
8.591242
8.626651
8.645062
8.671851
8.704031
8.723127
8.737705
8.750433
8.758621
8.772876
8.790050
8.794088
8.798091
8.816624
8.836301
8.852974
8.872307
8.887314
8.902466
8.924601
8.944691
8.960662
8.973784
8.984161
9.000123
9.019129
9.035000
9.054575
9.075913
9.094302
9.114171
9.135554
9.157538
9.176307
9.193220
9.220434
9.246620
9.257001
9.243476
9.202275
9.160007
9.147657
9.157229
9.171294
9.189745
9.198273
9.201042
9.214450
9.225220
9.233702
9.247974
9.262110
9.272097
9.275883
9.277719
9.278357
9.282812
9.289376
9.290788
9.292530
9.295274
9.297219
9.297555
9.290289
9.271941
9.255655];
ipc=[23.5993149
24.9374227
15.5668177
12.822998
8.720567
5.2893706
3.4198277
3.274742
3.664711
3.0744047
2.0693272
1.704983
1.3786582
1.3437517
2.6167173
4.7126735
6.3341827
7.3020707
6.5659735
4.3717219
3.4512364
4.7505574
5.6671666
5.6241466
5.5933745
4.7565382
3.7032831
3.6624086
3.8250059
3.2723449
3.3568247
3.7965188
3.5401286
2.7513284
2.5482882
2.2870368
2.0836978
2.6555146
3.3263235
3.3905589
3.2762151
2.8684465
2.1706005
2.3531933
2.5136208
2.3339758
2.0782414
1.8397957
1.7220838
2.3651145
3.4129652
3.8212567
3.9356373
3.6723844
3.11287
2.9099557
3.1280929
2.5837809
1.5138311
1.191679
1.286452
1.8904928
2.4851907
2.7408867
2.2572768
1.3839678
0.813925
0.88364
1.2723562
2.0426676
2.4161978
1.5967846
0.9763814
1.6120089
2.0962214
1.6427865
1.5391928
2.0537506
3.1496297
4.34142
4.3532191
2.2026746];