Skip to content
GitLab
Explore
Sign in
Register
Primary navigation
Search or go to…
Project
T
TZcode
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Wiki
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Snippets
Build
Pipelines
Jobs
Pipeline schedules
Artifacts
Deploy
Releases
Container registry
Model registry
Operate
Environments
Monitor
Incidents
Analyze
Value stream analytics
Contributor analytics
CI/CD analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
Contributions
TZcode
Commits
d19f2aeb
Commit
d19f2aeb
authored
13 years ago
by
Houtan Bastani
Browse files
Options
Downloads
Patches
Plain Diff
remove files of the form *.old
parent
c87b8fa4
Branches
Branches containing commit
No related tags found
No related merge requests found
Changes
2
Show whitespace changes
Inline
Side-by-side
Showing
2 changed files
MatlabFiles/ERRORS.OLD
+0
-91
0 additions, 91 deletions
MatlabFiles/ERRORS.OLD
MatlabFiles/IMPULSE.OLD
+0
-58
0 additions, 58 deletions
MatlabFiles/IMPULSE.OLD
with
0 additions
and
149 deletions
MatlabFiles/ERRORS.OLD
deleted
100755 → 0
+
0
−
91
View file @
c87b8fa4
function [vd,str,imf] = errors(Bh,swish,nn)
% Computing variance decompositions and impulse functions with
% [vd,str,imf] = errors(Bh,swish,nn)
% where imf and vd is of the same format as in RATS, that is to say:
% Column: nvar responses to 1st shock,
% nvar responses to 2nd shock, and so on.
% Row: steps of impulse responses.
% vd: variance decompositions
% str: standard errors of each variable, steps-by-nvar
% imf: impulse response functions
% Bh is the estimated reduced form coefficient in the form
% Y(T*nvar) = XB + U, X: T*k, B: k*nvar. The matrix
% form or dimension is the same as "Bh" from the function "sye";
% swish is the inv(A0) in the structural model A(L)y(t) = e(t).
% nn is the numbers of inputs [nvar,lags,# of impulse responses].
nvar = nn(1);
lags = nn(2);
imstep = nn(3); % number of steps for impulse responses
Ah = Bh';
% Row: nvar equations
% Column: 1st lag (with nvar variables) to lags (with nvar variables) + const = k.
imf = zeros(imstep,nvar*nvar);
vd = imf;
% Column: nvar responses to 1st shock, nvar responses to 2nd shock, and so on.
% Row: steps of impulse responses.
str = zeros(imstep,nvar); % initializing standard errors of each equation
M = zeros(nvar*(lags+1),nvar);
% Stack M0;M1;M2;...;Mlags
M(1:nvar,:) = swish;
Mtem = M(1:nvar,:); % temporary M -- impulse responses.
%
Mvd = Mtem.^2; % saved for the cumulative sum later
Mvds = (sum(Mvd'))';
str(1,:) = sqrt(Mvds'); % standard errors of each equation
Mvds = Mvds(:,ones(size(Mvds,1),1));
Mvdtem = (100.0*Mvd) ./ Mvds; % tempoary Mvd -- variance decompositions
% first or initial responses to
% one standard deviation shock (or forecast errors).
% Row: responses; Column: shocks
%
% * put in the form of "imf"
imf(1,:) = Mtem(:)';
vd(1,:) = Mvdtem(:)';
t = 1;
ims1 = min([imstep-1 lags]);
while t <= ims1
Mtem = zeros(nvar,nvar);
for k = 1:t
Mtem = Ah(:,nvar*(k-1)+1:nvar*k)*M(nvar*(t-k)+1:nvar*(t-k+1),:) + Mtem;
% Row: nvar equations, each for the nvar variables at tth lag
end
% ** impulse response functions
M(nvar*t+1:nvar*(t+1),:) = Mtem;
imf(t+1,:) = Mtem(:)';
% stack imf with each step, Row: 6 var to 1st shock, 6 var to 2nd shock, etc.
% ** variance decompositions
Mvd = Mvd + Mtem.^2; % saved for the cumulative sum later
Mvds = (sum(Mvd'))';
str(t+1,:) = sqrt(Mvds'); % standard errors of each equation
Mvds = Mvds(:,ones(size(Mvds,1),1));
Mvdtem = (100.0*Mvd) ./ Mvds; % tempoary Mvd -- variance decompositions
vd(t+1,:) = Mvdtem(:)';
% stack vd with each step, Row: 6 var to 1st shock, 6 var to 2nd shock, etc.
t= t+1;
end
for t = lags+1:imstep-1
M(1:nvar*lags,:) = M(nvar+1:nvar*(lags+1),:);
Mtem = zeros(nvar,nvar);
for k = 1:lags
Mtem = Ah(:,nvar*(k-1)+1:nvar*k)*M(nvar*(lags-k)+1:nvar*(lags-k+1),:) + Mtem;
% Row: nvar equations, each for the nvar variables at tth lag
end
% ** impulse response functions
M(nvar*lags+1:nvar*(lags+1),:) = Mtem;
imf(t+1,:) = Mtem(:)';
% stack imf with each step, Row: 6 var to 1st shock, 6 var to 2nd shock, etc.
% ** variance decompositions
Mvd = Mvd + Mtem.^2; % saved for the cumulative sum later
Mvds = (sum(Mvd'))';
str(t+1,:) = sqrt(Mvds'); % standard errors of each equation
Mvds = Mvds(:,ones(size(Mvds,1),1));
Mvdtem = (100.0*Mvd) ./ Mvds; % tempoary Mvd -- variance decompositions
vd(t+1,:) = Mvdtem(:)';
% stack vd with each step, Row: 6 var to 1st shock, 6 var to 2nd shock, etc.
end
\ No newline at end of file
This diff is collapsed.
Click to expand it.
MatlabFiles/IMPULSE.OLD
deleted
100755 → 0
+
0
−
58
View file @
c87b8fa4
function imf = impulse(Bh,swish,nn)
% Computing impulse functions with
% imf = impulse(Bh,swish,nn)
% where imf is in a format that is the SAME as in RATS.
% Column: nvar responses to 1st shock,
% nvar responses to 2nd shock, and so on.
% Row: steps of impulse responses.
% Bh is the estimated reduced form coefficient in the form
% Y(T*nvar) = XB + U, X: T*k, B: k*nvar. The matrix
% form or dimension is the same as "Bh" from the function "sye";
% swish is the inv(A0) in the structural model A(L)y(t) = e(t).
% nn is the numbers of inputs [nvar,lags,# of impulse responses].
nvar = nn(1);
lags = nn(2);
imstep = nn(3); % number of steps for impulse responses
Ah = Bh';
% Row: nvar equations
% Column: 1st lag (with nvar variables) to lags (with nvar variables) + const = k.
imf = zeros(imstep,nvar*nvar);
% Column: nvar responses to 1st shock, nvar responses to 2nd shock, and so on.
% Row: steps of impulse responses.
M = zeros(nvar*(lags+1),nvar);
% Stack M0;M1;M2;...;Mlags
M(1:nvar,:) = swish;
Mtem = M(1:nvar,:); % temporary M.
% first (initial) responses to 1 standard deviation shock. Row: responses; Column: shocks
% * put in the form of "imf"
imf(1,:) = Mtem(:)';
t = 1;
ims1 = min([imstep-1 lags]);
while t <= ims1
Mtem = zeros(nvar,nvar);
for k = 1:t
Mtem = Ah(:,nvar*(k-1)+1:nvar*k)*M(nvar*(t-k)+1:nvar*(t-k+1),:) + Mtem;
% Row: nvar equations, each for the nvar variables at tth lag
end
M(nvar*t+1:nvar*(t+1),:) = Mtem;
imf(t+1,:) = Mtem(:)';
% stack imf with each step, Row: 6 var to 1st shock, 6 var to 2nd shock, etc.
t= t+1;
end
for t = lags+1:imstep-1
M(1:nvar*lags,:) = M(nvar+1:nvar*(lags+1),:);
Mtem = zeros(nvar,nvar);
for k = 1:lags
Mtem = Ah(:,nvar*(k-1)+1:nvar*k)*M(nvar*(lags-k)+1:nvar*(lags-k+1),:) + Mtem;
% Row: nvar equations, each for the nvar variables at tth lag
end
M(nvar*lags+1:nvar*(lags+1),:) = Mtem;
imf(t+1,:) = Mtem(:)';
% stack imf with each step, Row: 6 var to 1st shock, 6 var to 2nd shock, etc.
end
\ No newline at end of file
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment