Merge branch 'master' of ssh://kirikou.dynare.org/srv/d_kirikou/git/dynare

m4/ax_mexopts.m4

-dnl Copyright (C) 2009-2011 Dynare Team
+dnl Copyright (C) 2009-2012 Dynare Team
 dnl
 dnl This file is part of Dynare.
 dnl
@@ -62,7 +62,7 @@ case ${MATLAB_ARCH} in
     ax_mexopts_ok="yes"
     ;;
   maci | maci64)
-    SDKROOT='/Developer/SDKs/MacOSX10.6.sdk'
+    SDKROOT='/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.6.sdk'
     MACOSX_DEPLOYMENT_TARGET='10.6'
     if test "${MATLAB_ARCH}" = "maci"; then
         ARCHS='i386'

matlab/DsgeVarLikelihood.m

@@ -35,7 +35,9 @@ function [fval,grad,hess,exit_flag,info,PHI,SIGMAu,iXX,prior] = DsgeVarLikelihoo
 % along with Dynare.  If not, see <http://www.gnu.org/licenses/>.
 
 % Declaration of the persistent variables.
-persistent penalty dsge_prior_weight_idx
+persistent dsge_prior_weight_idx
+
+penalty = BayesInfo.penalty;
 
 grad=[];
 hess=[];
@@ -46,16 +48,6 @@ SIGMAu = [];
 iXX = [];
 prior = [];
 
-% Initialization of the penalty
-if ~nargin || isempty(penalty)
-    penalty = 1e8;
-    if ~nargin, return, end
-end
-if nargin==1
-    penalty = xparam1;
-    return
-end
-
 % Initialization of of the index for parameter dsge_prior_weight in Model.params.
 if isempty(dsge_prior_weight_idx)
     dsge_prior_weight_idx = strmatch('dsge_prior_weight',Model.param_names);

matlab/cmaes.m

@@ -644,11 +644,11 @@ else % flgresume
   end
     
   % initialize random number generator
-  if ischar(opts.Seed)
-    randn('state', eval(opts.Seed));     % random number generator state
-  else
-    randn('state', opts.Seed);
-  end
+% $$$   if ischar(opts.Seed)
+% $$$     randn('state', eval(opts.Seed));     % random number generator state
+% $$$   else
+% $$$     randn('state', opts.Seed);
+% $$$   end
   %qqq
 %  load(opts.SaveFilename, 'startseed');
 %  randn('state', startseed);

matlab/dsge_likelihood.m

@@ -130,22 +130,7 @@ function [fval,DLIK,Hess,exit_flag,ys,trend_coeff,info,Model,DynareOptions,Bayes
 
 % AUTHOR(S) stephane DOT adjemian AT univ DASH lemans DOT FR
 
-% Declaration of the penalty as a persistent variable.
-
-% Persistent variable 'penalty' is used to compute an endogenous penalty to
-% the value 'fval' when various conditions are encountered. These conditions
-% set also 'exit_flag' equal to 0 instead of 1.  It is only when
-% dsge_likelihood() is called by an optimizer called by
-% dynare_estimation_1() that 'exit_flag' is ignored and penalized 'fval' is
-% actually used.
-% In that case, 'penalty' is properly initialized, at the very end of the
-% present function, by a call to dsge_likelihood() made in
-% initial_estimation_checks(). If a condition triggers exit_flag ==
-% 0, initial_estimation_checks() triggers an error.
-% In summary, an initial call to the present function, without triggering
-% any condition, guarantees that 'penalty' is properly initialized when needed.
-
-persistent penalty
+penalty = BayesInfo.penalty;
 
 % Initialization of the returned variables and others...
 fval        = [];
@@ -568,16 +553,16 @@ if analytic_derivation,
         if full_Hess
         DTj = DT(:,:,j+offset);
         DPj = dum;
-        for i=1:j,
-            DTi = DT(:,:,i+offset);
-            DPi = DP(:,:,i+offset);
-            D2Tij = D2T(:,:,i,j);
-            D2Omij = D2Om(:,:,i,j);
+        for i=1:j+offset,
+            DTi = DT(:,:,i);
+            DPi = DP(:,:,i);
+            D2Tij = D2T(:,:,i,j+offset);
+            D2Omij = D2Om(:,:,i,j+offset);
             tmp = D2Tij*Pstar*T' + T*Pstar*D2Tij' + DTi*DPj*T' + DTj*DPi*T' + T*DPj*DTi' + T*DPi*DTj' + DTi*Pstar*DTj' + DTj*Pstar*DTi' + D2Omij;
             dum = lyapunov_symm(T,tmp,DynareOptions.qz_criterium,DynareOptions.lyapunov_complex_threshold);
             dum(abs(dum)<1.e-12) = 0;
-            D2P(:,:,i+offset,j+offset) = dum;
-            D2P(:,:,j+offset,i+offset) = dum;
+            D2P(:,:,i,j+offset) = dum;
+            D2P(:,:,j+offset,i) = dum;
         end
         end
     end
@@ -771,9 +756,6 @@ end
 % Update DynareOptions.kalman_algo.
 DynareOptions.kalman_algo = kalman_algo;
 
-% Update the penalty.
-penalty = fval;
-
 if analytic_derivation==0 && nargout==2,
     lik=lik(start:end,:);
     DLIK=[-lnprior; lik(:)];

matlab/k_order_pert.m

@@ -159,7 +159,7 @@ for i=1:n1
             xx = x(:,m);
             y(:,(i-1)*n1*n2+(j-1)*n2+k) = xx;
             if j ~= i
-                y(:,(j-1)*n1*n2+(i-1)*n2+i) = xx;
+                y(:,(j-1)*n1*n2+(i-1)*n2+k) = xx;
             end
         end
     end

matlab/k_order_perturbation.m

+% [err, g_0, g_1, g_2, g_3, derivs] = k_order_perturbation(dr,DynareModel,DynareOptions)
+% computes a k_order_petrubation solution for k=1,2,3
+%
+% INPUTS
+% dr:            struct   describing the reduced form solution of the model.
+% DynareModel:   struct   jobs's parameters
+% DynareOptions: struct   job's options
+%
+% OUTPUTS
+% err:           double   err code (currently unused)
+% g_0:           vector   dynare++ output. Constant effect of future volatility (in
+%                         dr.order_var order) on the decision rule
+%                         (=ghs2/2). Contains zero when order of
+%                         approximation is 1.
+% g_1:           matrix   dynare++ output. First order Taylor coefficients
+%                         of decision rule. When order of approximation
+%                         is 3, the. Contains both the effect of
+%                         state endogenous variable and shocks. The rows
+%                         are in dr.order_var order. The columns are in
+%                         dr.order_var order of state endogenous
+%                         variables and shocks
+% g_2:           matrix   dynare++ output. Second order Taylor coefficients of decision
+%                         rule. Contains both the effect of state endogenous
+%                         variable and shocks. The rows are in dr.order_var
+%                         order. Each row corresponds to the vectorized
+%                         version of the lower triangle of the Hessian
+%                         matrix. The Taylor coefficient (1/2) is
+%                         included. The columns of the Hessian matrix are in
+%                         dr.order_var order of state endogenous variables
+%                         and shocks
+% g_3:           matrix   dynare++ output. Third order Taylor coefficients of decision
+%                         rule. Contains both the effect of state endogenous
+%                         variable and shocks. The rows are in dr.order_var
+%                         order. Each row corresponds to the vectorized
+%                         version of the 3rd order derivatives tensor where each
+%                         combination of variables appears only once.
+%                         The Taylor coefficient (1/6) is
+%                         included. Inside the tensor, the variables are in
+%                         dr.order_var order of state endogenous variables
+%                         and shocks 
+% derivs        struct    contains the original derivatives of the
+%                         decision function (ghx, ghu, ghxx, ghxu, ghuu,
+%                         ghs2, ghxxx, ghxxu, ghxuu,ghuuu, ghxss, ghuss),
+%                         keeping the effect of future volatility
+%                         separate (in  ghs2, ghxss and ghuss). The
+%                         derivatives matrices contain full versions of
+%                         the Hessian matrices and 3rd order
+%                         tensor. Symmetric derivatives are repeated. The
+%                         Taylor coefficients (1/2 and 1/6) aren't
+%                         included.
+% k_order_peturbation is a compiled MEX function. It's source code is in
+% dynare/mex/sources/k_order_perturbation.cc and it uses code provided by
+% dynare++

matlab/metropolis_hastings_initialization.m

@@ -180,9 +180,6 @@ if ~options_.load_mh_file && ~options_.mh_recover
     % separate initializaton for each chain
     JSUM = 0;
     for j=1:nblck,
-        JSUM  = JSUM + sum(100*clock);
-        randn('state',JSUM);
-        rand('state',JSUM);
         record.Seeds(j).Normal = randn('state');
         record.Seeds(j).Unifor = rand('state');
     end

mex/build/matlab/configure.ac

 dnl Process this file with autoconf to produce a configure script.
 
-dnl Copyright (C) 2009-2011 Dynare Team
+dnl Copyright (C) 2009-2012 Dynare Team
 dnl
 dnl This file is part of Dynare.
 dnl
@@ -17,7 +17,7 @@ dnl
 dnl You should have received a copy of the GNU General Public License
 dnl along with Dynare.  If not, see <http://www.gnu.org/licenses/>.
 
-AC_PREREQ([2.61])
+AC_PREREQ([2.62])
 AC_INIT([dynare], [4.4-unstable])
 AC_CONFIG_SRCDIR([configure.ac])
 AM_INIT_AUTOMAKE([-Wall -Werror foreign])

mex/build/octave/configure.ac

 dnl Process this file with autoconf to produce a configure script.
 
-dnl Copyright (C) 2009-2011 Dynare Team
+dnl Copyright (C) 2009-2012 Dynare Team
 dnl
 dnl This file is part of Dynare.
 dnl
@@ -17,7 +17,7 @@ dnl
 dnl You should have received a copy of the GNU General Public License
 dnl along with Dynare.  If not, see <http://www.gnu.org/licenses/>.
 
-AC_PREREQ([2.61])
+AC_PREREQ([2.62])
 AC_INIT([dynare], [4.4-unstable])
 AC_CONFIG_SRCDIR([configure.ac])
 AM_INIT_AUTOMAKE([-Wall -Werror foreign])

mex/sources/k_order_perturbation/k_order_perturbation.cc

@@ -260,8 +260,6 @@ extern "C" {
         // run stochastic steady
         app.walkStochSteady();
 
-	app.get_rule_ders()->print();
-
         /* Write derivative outputs into memory map */
         map<string, ConstTwoDMatrix> mm;
         app.getFoldDecisionRule().writeMMap(mm, string());