Documentation of idsimsd

Global Index (short | long) | Local contents | Local Index (short | long)

Function Synopsis

idsimsd(u,th,n,noise,ky)

Help text

IDSIMSD Illustrates the uncertainty in simulated model responses.
   IDSIMSD(U,TH)

   U is a column vector (matrix) containing the input(s).
   TH is a model given in the THETA format (See also THETA).

   10 random models are created, consistent with the covariance informa-
   tion in TH, and the responses of each of these models to U are plotted
   in the same diagram.

   The number 10 can be changed to N by IDSIMSD(U,TH,N).

   With IDSIMSD(U,TH,N,'noise',KY), additive noise (e) is added to the
   simulation in accordance with the noise model of TH.
   KY denotes the output numbers to be plotted (default all).

   See also IDSIM.

Cross-Reference Information

Listing of function idsimsd

function idsimsd(u,th,n,noise,ky)

%   L.Ljung 7-8-87
%   Copyright (c) 1986-98 by The MathWorks, Inc.
%   $Revision: 2.3 $  $Date: 1997/12/02 03:42:53 $

if nargin < 2
   disp('Usage: IDSIMSD(INPUT,MODEL)')
   disp('       IDISMSD(INPUT,MODEL,ADD_NOISE,OUTPUTS)')
   disp('       ADD_NOISE one of ''no_noise'', ''noise''.')
   return
end

if isthss(th),ny=th(1,4);else ny=1;end
if nargin<5,ky=[];end
if nargin <4,noise=[];end
if nargin<3,n=[];end,
if isempty(ky),ky=1:ny;end
if isempty(noise),noise='nonoise';end
if isempty(n),n=10;end

nu=th(1,3);Tsamp=abs(th(1,2));
[Nc,d]=getncap(th);
[N,nz]=size(u);
if nu~=nz, error('The input matrix U has not the correct number of columns!'),
return,end
[par,P,lam]=th2par(th);
if size(P)==0; disp('No covariance information given in TH'),P=zeros(d,d);end
P=chol(P);
u1=u;
for kk=ky
if noise(3)=='i', u1=[u randn(N,ny)];end %corr 9007
yh=idsim(u1,th);yh=yh(:,kk);
ndu=length(yh);y1=max(yh);y2=min(yh);
y12=y1-y2;y1=y1+0.2*y12;y2=y2-0.2*y12;
plot([1:ndu]*Tsamp,yh),axis([Tsamp ndu*Tsamp y2 y1]);hold on;
title(['Output number ',int2str(kk)])
u1=u;
for k=1:n-1
        th1=th; th1(3,1:d)=par+randn(1,d)*P;
        if noise(3)=='i', u1=[u randn(N,ny)];end
        yh=idsim(u1,th1);
        plot([1:ndu]*Tsamp,yh(:,kk))
end
hold off
if kk<ky(length(ky)),pause,end
end
set(gcf,'NextPlot','replace');