yuwa.m

时间序列工具箱

function [arp] = yuwa(ACF);
% use DURLEV.M, this function will be refused in future versions
%
% function  [arp] = yuwa(ACF);
% estimates AR(p) model parameter by solving the
% Yule-Walker equations
% ACF	Autocovariance function [0..P]
% arp	autoregressive model parameter	

%	Version 2.40
%	last revision 12.05.1998
%	Copyright (c) 1996-1998 by Alois Schloegl
%	e-mail: [e-mail]	

% This library is free software; you can redistribute it and/or
% modify it under the terms of the GNU Library General Public
% License as published by the Free Software Foundation; either
% Version 2 of the License, or (at your option) any later version.
%
% This library 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
% Library General Public License for more details.
%
% You should have received a copy of the GNU Library General Public
% License along with this library; if not, write to the
% Free Software Foundation, Inc., 59 Temple Place - Suite 330,
% Boston, MA  02111-1307, USA.


[p,nc]=size(ACF);
%if p<nc	ACF=ACF'; p=nc; end;

arp=durlev(ACF)';
return;

p=p-1;
if (p>195),     % use higher overhead, but asymptotically faster algorithm
	%	Levinson-Durbin Algorithm
	arp=zeros(p,1);
	arp=ACF(1);	% Inititialization
	for K=1:p-1,
		arp(K+1)=(ACF(K+1)-arp(1:K)*ACF(K:-1:1))/(1-arp(1:K)*ACF(1:K));
    		arp(1:K)=PHI(1:K)-arp(K+1)*arp(K:-1:1);	
	end;
else   % use the good old \ command
	arp = toeplitz(ACF(1:p))\ACF(2:p+1);
end;
return;