g13acf
g13acf
© Numerical Algorithms Group, 2002.
Purpose
G13ACF Univariate time series, partial autocorrelations from
autocorrelations
Synopsis
[p,v,ar,nvl,ifail] = g13acf(r,nl<,ifail>)
Description
The data consist of values of autocorrelation coefficients
r ,r ,...,r , relating to lags 1,2,...,K. These will generally
1 2 K
(but not necessarily) be sample values such as may be obtained
from a time series x using G13ABF.
t
The partial autocorrelation coefficient at lag l may be
identified with the parameter p in the autoregression
l,l
x =c +p x +p x +...+p x +e
t l l,1 t-1 l,2 t-2 l,l t-l l,t
where e is the predictor error.
l,t
The first subscript l of p and e emphasises the fact that
l,l l,t
the parameters will in general alter as further terms are
introduced into the equation (i.e., as l is increased).
The parameters are determined from the autocorrelation
coefficients by the Yule-Walker equations
r =p r +p r +...+p r , i=1,2,...,l
i l,1 i-1 l,2 i-2 l,l i-l
taking r =r when j<0, and r =1.
j |j| 0
The predictor error variance ratio v =var(e )/var(x ) is
l l,t t
defined by v =1-p r -p r -...-p r .
l l,1 1 l,2 2 l,l l
The above sets of equations are solved by a recursive method (the
Durbin-Levinson algorithm). The recursive cycle applied for
l=1,2,...,(L-1), where L is the number of partial autocorrelation
coefficients required, is initialised by setting p =r and
1,1 1
2
v =1-r .
1 1
Then
p = (r -p r -p r -...-p r )/v
l+1,l+1 l+1 l,1 l l,2 l-1 l,l 1 l
p = p -p p , j=1,2,...,l
l+1,j l,j l+1,l+1 l,l+1-j
v = v (1-p )(1+p ).
l+1 l l+1,l+1 l+1,l+1
If the condition |p |>=1 occurs, say when l=l , it indicates
l,l 0
that the supplied autocorrelation coefficients do not form a
positive-definite sequence, and the recursion is
not continued. The autoregressive parameters are overwritten at
each recursive step, so that upon completion the only available
values are p , for j=1,2,...,L or p if the recursion has
Lj l -1,j
0
been prematurely halted.
Parameters
g13acf
Required Input Arguments:
r (:) real
nl integer
Optional Input Arguments: <Default>
ifail integer -1
Output Arguments:
p (nl) real
v (nl) real
ar (nl) real
nvl integer
ifail integer