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

Merge remote branch 'ratto/master'

time-shift
Sébastien Villemot 2011-02-14 14:29:39 +01:00
parent 9eb4f5a067 90e5f489b2
commit 94ca528604
9 changed files with 90 additions and 55 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
matlab/DsgeLikelihood_hh.m
View File

@ -266,4 +266,6 @@ end
lnprior = priordens(xparam1,bayestopt_.pshape,bayestopt_.p6,bayestopt_.p7,bayestopt_.p3,bayestopt_.p4);
fval = (likelihood-lnprior);
options_.kalman_algo = kalman_algo;
llik=[-lnprior; lik(start:end)];
lik=lik(start:end,:);
llik=[-lnprior; lik(:)];
% llik=[-lnprior; lik(start:end)];

2
matlab/ReshapeMatFiles.m
View File

@ -115,7 +115,7 @@ switch TYPEarray
eval(['STOCK_' CAPtype ' = zeros(NumberOfPeriodsPerTYPEfiles,TYPEsize(2),TYPEsize(3),B);'])
for f2 = 1:NumberOfTYPEfiles
load([MhDirectoryName M_.fname '_' type int2str(f2) '.mat']);
eval(['STOCK_' CAPtype '(:,:,:,idx+1:idx+size(stock_' type ',4))=stock_' ...
eval(['STOCK_' CAPtype '(:,:,1:+size(stock_' type ',3),idx+1:idx+size(stock_' type ',4))=stock_' ...
type '(jdx+1:jdx+NumberOfPeriodsPerTYPEfiles,:,:,:);'])
eval(['idx = idx + size(stock_' type ',4);'])
end

18
matlab/kalman/likelihood/univariate_diffuse_kalman_filter.m
View File

@ -1,4 +1,4 @@
function [LIK, lik] = univariate_diffuse_kalman_filter(T,R,Q,H,Pinf,Pstar,Y,start,Z,kalman_tol,riccati_tol,data_index,number_of_observations,no_more_missing_observations)
function [LIK, llik] = univariate_diffuse_kalman_filter(T,R,Q,H,Pinf,Pstar,Y,start,Z,kalman_tol,riccati_tol,data_index,number_of_observations,no_more_missing_observations)
% Computes the likelihood of a stationnary state space model (univariate approach).
%
% INPUTS
@ -52,6 +52,7 @@ a = zeros(mm,1);
QQ = R*Q*transpose(R);
t = 0;
lik = zeros(smpl,1);
llik=zeros(smpl,pp);
notsteady = 1;
crit = 1.e-6;
newRank = rank(Pinf,crit);
@ -74,10 +75,12 @@ while newRank && (t<smpl)
Pstar = Pstar + Kinf*(Kinf_Finf'*(Fstar/Finf)) - Kstar*Kinf_Finf' ...
- Kinf_Finf*Kstar';
Pinf = Pinf - Kinf*Kinf_Finf';
lik(t) = lik(t) + log(Finf) + l2pi;
llik(t,i) = log(Finf) + l2pi;
lik(t) = lik(t) + llik(t,i);
elseif Fstar>kalman_tol
lik(t) = lik(t) + log(Fstar) + prediction_error* ...
llik(t,i) = log(Fstar) + prediction_error* ...
prediction_error/Fstar + l2pi;
lik(t) = lik(t) + llik(t,i);
a = a + Kstar*(prediction_error/Fstar);
Pstar = Pstar - Kstar*(Kstar'/Fstar);
end
@ -116,8 +119,9 @@ while notsteady && (t<smpl)
if Fi > kalman_tol
a = a + Ki*(prediction_error/Fi);
Pstar = Pstar - Ki*(Ki'/Fi);
lik(t) = lik(t) + log(Fi) + prediction_error*prediction_error/Fi ...
llik(t,i) = log(Fi) + prediction_error*prediction_error/Fi ...
+ l2pi;
lik(t) = lik(t) + llik(t,i);
end
end
a = T*a;
@ -138,13 +142,15 @@ while t < smpl
Ki = Pstar*Zi';
a = a + Ki*prediction_error/Fi;
Pstar = Pstar - Ki*Ki'/Fi;
lik(t) = lik(t) + log(Fi) + prediction_error*prediction_error/Fi ...
llik(t,i) = log(Fi) + prediction_error*prediction_error/Fi ...
+ l2pi;
lik(t) = lik(t) + llik(t,i);
end
end
a = T*a;
end
lik = lik/2;
llik = llik/2;
LIK = sum(lik(start:end));
LIK = sum(lik(start:end));

16
matlab/kalman/likelihood/univariate_diffuse_kalman_filter_corr.m
View File

@ -1,4 +1,4 @@
function [LIK, lik] = univariate_diffuse_kalman_filter_corr(T,R,Q,H,Pinf,Pstar,Y,start,Z,kalman_tol,riccati_tol,data_index,number_of_observations,no_more_missing_observations)
function [LIK, llik] = univariate_diffuse_kalman_filter_corr(T,R,Q,H,Pinf,Pstar,Y,start,Z,kalman_tol,riccati_tol,data_index,number_of_observations,no_more_missing_observations)
% Computes the likelihood of a stationnary state space model (univariate
% approach with correlated errors).
%
@ -52,6 +52,7 @@ a = zeros(mm,1);
QQ = R*Q*transpose(R);
t = 0;
lik = zeros(smpl,1);
llik = zeros(smpl,pp);
notsteady = 1;
crit = 1.e-6;
newRank = rank(Pinf,crit);
@ -104,7 +105,8 @@ while newRank && (t<smpl)
a = a + Kinf*(prediction_error/Finf);
Pstar = Pstar + Kinf*(Kinf'*(Fstar/(Finf*Finf))) - (Kstar*Kinf'+Kinf*Kstar')/Finf;
Pinf = Pinf - Kinf*(Kinf'/Finf);
lik(t) = lik(t) + log(Finf) + l2pi;
llik(t,i) = log(Finf) + l2pi;
lik(t) = lik(t) + llik(t,i);
if ~isempty(options_.diffuse_d)
newRank = (icc<options_.diffuse_d);
if newRank && (any(diag(Za*Pinf*Za')>kalman_tol)==0 & rank(Pinf,crit)==0);
@ -125,8 +127,9 @@ while newRank && (t<smpl)
end
end
elseif Fstar>kalman_tol
lik(t) = lik(t) + log(Fstar) + prediction_error* ...
llik(t,i) = log(Fstar) + prediction_error* ...
prediction_error/Fstar + l2pi;
lik(t) = lik(t) + llik(t,i);
a = a + Kstar*prediction_error/Fstar;
Pstar = Pstar - Kstar*Kstar'/Fstar;
end
@ -165,8 +168,9 @@ while notsteady && (t<smpl)
Ki = Pstar*Zi';
a = a + Ki*prediction_error/Fi;
Pstar = Pstar - Ki*Ki'/Fi;
lik(t) = lik(t) + log(Fi) + prediction_error*prediction_error/Fi ...
llik(t,i) = log(Fi) + prediction_error*prediction_error/Fi ...
+ l2pi;
lik(t) = lik(t) + llik(t,i);
end
end
a = T*a;
@ -187,13 +191,15 @@ while t < smpl
Ki = Pstar*Zi';
a = a + Ki*prediction_error/Fi;
Pstar = Pstar - Ki*Ki'/Fi;
lik(t) = lik(t) + log(Fi) + prediction_error*prediction_error/Fi ...
llik(t,i) = log(Fi) + prediction_error*prediction_error/Fi ...
+ l2pi;
lik(t) = lik(t) + llik(t,i);
end
end
a = T*a;
end
lik = lik/2;
llik = lik/2;
LIK = sum(lik(start:end));

12
matlab/kalman/likelihood/univariate_kalman_filter.m
View File

@ -1,4 +1,4 @@
function [LIK, lik] = univariate_kalman_filter(T,R,Q,H,P,Y,start,mf,kalman_tol,riccati_tol,data_index,number_of_observations,no_more_missing_observations)
function [LIK, llik] = univariate_kalman_filter(T,R,Q,H,P,Y,start,mf,kalman_tol,riccati_tol,data_index,number_of_observations,no_more_missing_observations)
% Computes the likelihood of a stationnary state space model (univariate approach).
%
% INPUTS
@ -18,7 +18,7 @@ function [LIK, lik] = univariate_kalman_filter(T,R,Q,H,P,Y,start,mf,kalman_tol,r
%
% OUTPUTS
% LIK [double] scalar, MINUS loglikelihood
% lik [double] vector, density of observations in each period.
% llik [double] vector, density of observations in each period.
%
% REFERENCES
% See "Filtering and Smoothing of State Vector for Diffuse State Space
@ -52,6 +52,7 @@ a = zeros(mm,1); % Initial condition of the state
QQ = R*Q*transpose(R);
t = 0;
lik = zeros(smpl,1);
llik = zeros(smpl,pp);
notsteady = 1;
l2pi = log(2*pi);
@ -67,8 +68,9 @@ while notsteady && t<smpl
Ki = P(:,MF(i))/Fi;
a = a + Ki*prediction_error;
P = P - (Fi*Ki)*transpose(Ki);
lik(t) = lik(t) + log(Fi) + prediction_error*prediction_error/Fi ...
llik(t,i) = log(Fi) + prediction_error*prediction_error/Fi ...
+ l2pi;
lik(t) = lik(t) + llik(t,i);
end
end
a = T*a;
@ -89,13 +91,15 @@ while t < smpl
Ki = PP(:,mf(i))/Fi;
a = a + Ki*prediction_error;
PP = PP - (Fi*Ki)*transpose(Ki);
lik(t) = lik(t) + log(Fi) + prediction_error*prediction_error/Fi ...
llik(t,i) = log(Fi) + prediction_error*prediction_error/Fi ...
+ l2pi;
lik(t) = lik(t) + llik(t,i);
end
end
a = T*a;
end
lik = lik/2;
llik = llik/2;
LIK = sum(lik(start:end));

12
matlab/kalman/likelihood/univariate_kalman_filter_corr.m
View File

@ -1,4 +1,4 @@
function [LIK, lik] = ...
function [LIK, llik] = ...
univariate_kalman_filter_corr(T,R,Q,H,P,Y,start,mf,kalman_tol,riccati_tol,data_index,number_of_observations,no_more_missing_observations)
% Computes the likelihood of a stationnary state space model (univariate
% approach + correlated measurement errors).
@ -20,7 +20,7 @@ function [LIK, lik] = ...
%
% OUTPUTS
% LIK [double] scalar, MINUS loglikelihood
% lik [double] vector, density of observations in each period.
% llik [double] vector, density of observations in each period.
%
% REFERENCES
% See "Filtering and Smoothing of State Vector for Diffuse State Space
@ -54,6 +54,7 @@ smpl = size(Y,2); % Number of periods in the dataset.
a = zeros(mm+pp,1); % Initial condition of the state vector.
t = 0;
lik = zeros(smpl,1);
llik = zeros(smpl,pp);
notsteady = 1;
TT = zeros(mm+pp);
@ -82,8 +83,9 @@ while notsteady && t<smpl
prediction_error = Y(d_index(i),t) - a(MF(i)) - a( mm+i );
Fi = PP(MF(i),MF(i)) + PP(mm+i,mm+i);
if Fi > kalman_tol
lik(t) = lik(t) + log(Fi) + prediction_error*prediction_error/Fi ...
llik(t,i) = log(Fi) + prediction_error*prediction_error/Fi ...
+ l2pi;
lik(t) = lik(t) + llik(t,i);
Ki = sum(PP(:,[MF(i) mm+i]),2)/Fi;
a = a + Ki*prediction_error;
PP = PP - (Ki*Fi)*transpose(Ki);
@ -111,8 +113,9 @@ while t < smpl
Ki = ( PPPP(:,mf(i)) + PPPP(:,mm+i) )/Fi;
a = a + Ki*prediction_error;
PPPP = PPPP - (Fi*Ki)*transpose(Ki);
lik(t) = lik(t) + log(Fi) + prediction_error*prediction_error/Fi ...
llik(t,i) = log(Fi) + prediction_error*prediction_error/Fi ...
+ l2pi;
lik(t) = lik(t) + llik(t,i);
end
end
a(1:mm) = T*a(1:mm);
@ -120,5 +123,6 @@ while t < smpl
end
lik = lik/2;
llik = llik/2;
LIK = sum(lik(start:end));

14
matlab/mr_gstep.m
View File

@ -27,7 +27,8 @@ n=size(x,1);
if init,
gstep_ = options_.gstep;
h1=max(abs(x),sqrt(gstep_)*ones(n,1))*eps^(1/4);
% h1=max(abs(x),sqrt(gstep_)*ones(n,1))*eps^(1/4);
h1=options_.gradient_epsilon*ones(n,1);
return
end
if nargin<4,
@ -71,7 +72,8 @@ while i<n,
icount = 0;
h0=h1(i);
while (abs(dx(it))<0.5*htol | abs(dx(it))>(2*htol)) & icount<10 & ic==0,
% while (abs(dx(it))<0.5*htol | abs(dx(it))>(2*htol)) & icount<10 & ic==0,
while (abs(dx(it))<0.5*htol) & icount<10 & ic==0,
%while abs(dx(it))<0.5*htol & icount< 10 & ic==0,
icount=icount+1;
if abs(dx(it)) ~= 0,
@ -86,10 +88,10 @@ while i<n,
% ic=1;
% end
end
if abs(dx(it))>(2*htol),
h1(i)= htol/abs(dx(it))*h1(i);
xh1(i)=x(i)+h1(i);
end
% if abs(dx(it))>(2*htol),
% h1(i)= htol/abs(dx(it))*h1(i);
% xh1(i)=x(i)+h1(i);
% end
try
fx = feval(func,xh1,varargin{:});
catch

52
matlab/mr_hessian.m
View File

@ -50,7 +50,8 @@ if init,
htol = 1.e-4;
%h1=max(abs(x),gstep_*ones(n,1))*eps^(1/3);
%h1=max(abs(x),sqrt(gstep_)*ones(n,1))*eps^(1/6);
h1=max(abs(x),sqrt(gstep_)*ones(n,1))*eps^(1/4);
% h1=max(abs(x),sqrt(gstep_)*ones(n,1))*eps^(1/4);
h1=options_.gradient_epsilon*ones(n,1);
return,
end
func = str2func(func);
@ -102,7 +103,8 @@ while i<n,
icount = 0;
h0=h1(i);
while (abs(dx(it))<0.5*htol | abs(dx(it))>(2*htol)) & icount<10 & ic==0,
% while (abs(dx(it))<0.5*htol | abs(dx(it))>(2*htol)) & icount<10 & ic==0,
while (abs(dx(it))<0.5*htol) & icount<10 & ic==0,
%while abs(dx(it))<0.5*htol & icount< 10 & ic==0,
icount=icount+1;
%if abs(dx(it)) ~= 0,
@ -127,21 +129,21 @@ while i<n,
fx=1.e8;
end
end
if abs(dx(it))>(2*htol),
h1(i)= htol/abs(dx(it))*h1(i);
xh1(i)=x(i)+h1(i);
try
[fx, ffx]=feval(func,xh1,varargin{:});
catch
fx=1.e8;
end
while (fx-f0)==0,
h1(i)= h1(i)*2;
xh1(i)=x(i)+h1(i);
[fx, ffx]=feval(func,xh1,varargin{:});
ic=1;
end
end
% if abs(dx(it))>(2*htol),
% h1(i)= htol/abs(dx(it))*h1(i);
% xh1(i)=x(i)+h1(i);
% try
% [fx, ffx]=feval(func,xh1,varargin{:});
% catch
% fx=1.e8;
% end
% while (fx-f0)==0,
% h1(i)= h1(i)*2;
% xh1(i)=x(i)+h1(i);
% [fx, ffx]=feval(func,xh1,varargin{:});
% ic=1;
% end
% end
it=it+1;
dx(it)=(fx-f0);
h0(it)=h1(i);
@ -171,7 +173,7 @@ while i<n,
else
ff1=ffx;
end
if hflag, % two point based derivatives
% if hflag, % two point based derivatives
xh1(i)=x(i)-h1(i);
% c=mr_nlincon(xh1,varargin{:});
% ic=0;
@ -193,9 +195,9 @@ while i<n,
% [f1(:,i), ff1]=feval(func,xh1,varargin{:});
% end
ggh(:,i)=(ff1-ff_1)./(2.*h1(i));
else
ggh(:,i)=(ff1-ff0)./h1(i);
end
% else
% ggh(:,i)=(ff1-ff0)./h1(i);
% end
xh1(i)=x(i);
if hcheck & htol<1,
htol=min(1,max(min(abs(dx))*2,htol*10));
@ -209,11 +211,11 @@ h_1=h1;
xh1=x;
xh_1=xh1;
if hflag,
% if hflag,
gg=(f1'-f_1')./(2.*h1);
else
gg=(f1'-f0)./h1;
end
% else
% gg=(f1'-f0)./h1;
% end
if hflag==2,
gg=(f1'-f_1')./(2.*h1);

15
matlab/newrat.m
View File

@ -82,6 +82,7 @@ else
hhg=hh;
igg=inv(hh);
end
H = igg;
disp(['Gradient norm ',num2str(norm(gg))])
ee=eig(hh);
disp(['Minimum Hessian eigenvalue ',num2str(min(ee))])
@ -105,7 +106,7 @@ while norm(gg)>gtol & check==0 & jit<nit,
bayestopt_.penalty = fval0(icount);
disp([' '])
disp(['Iteration ',num2str(icount)])
[fval x0 fc retcode] = csminit(func0,xparam1,fval0(icount),gg,0,igg,varargin{:});
[fval x0 fc retcode] = csminit(func0,xparam1,fval0(icount),gg,0,H,varargin{:});
if igrad,
[fval1 x01 fc retcode1] = csminit(func0,x0,fval,gg,0,inx,varargin{:});
if (fval-fval1)>1, %(fval0(icount)-fval),
@ -116,7 +117,7 @@ while norm(gg)>gtol & check==0 & jit<nit,
fval=fval1;
x0=x01;
end
if (fval0(icount)-fval)<1.e-2*(gg'*(igg*gg))/2 & igibbs,
if (fval0(icount)-fval)<1.e-2*(gg'*(H*gg))/2 & igibbs,
if length(find(ig))<nx,
ggx=ggx*0;
ggx(find(ig))=gg(find(ig));
@ -126,6 +127,9 @@ while norm(gg)>gtol & check==0 & jit<nit,
[fvala x0 fc retcode] = csminit(func0,x0,fval,ggx,0,iggx,varargin{:});
end
[fvala, x0, ig] = mr_gstep(0,x0,func0,htol,varargin{:});
% if length(find(ig))==0,
% [fvala, x0, ig] = mr_gstep(0,x0,func0,htol/10,varargin{:});
% end
nig=[nig ig];
if (fval-fvala)<gibbstol*(fval0(icount)-fval),
igibbs=0;
@ -160,6 +164,8 @@ while norm(gg)>gtol & check==0 & jit<nit,
x(:,icount+1)=xparam1;
fval0(icount+1)=fval;
%if (fval0(icount)-fval)<ftol*ftol & flagg==1;,
mr_gstep(1,x);
mr_hessian(1,x);
if (fval0(icount)-fval)<ftol,
disp('No further improvement is possible!')
check=1;
@ -228,6 +234,7 @@ while norm(gg)>gtol & check==0 & jit<nit,
end
end
end
disp(['Gradient norm ',num2str(norm(gg))])
ee=eig(hh);
disp(['Minimum Hessian eigenvalue ',num2str(min(ee))])
@ -237,7 +244,9 @@ while norm(gg)>gtol & check==0 & jit<nit,
disp(['Elapsed time for iteration ',num2str(t),' s.'])
g(:,icount+1)=gg;
save m1.mat x hh g hhg igg fval0 nig
% H = bfgsi(H,g(:,end)-g(:,end-1),x(:,end)-x(:,end-1));
H = igg;
save m1.mat x hh g hhg igg fval0 nig H
end
end