Example 12.1 (AR(1))
The AR(1) process from Example
10.1 with
has the
characteristic equation
. The explicit solution
and
occurs exactly when
. The
inverse filter of
is thus
and the
MA(
) representation of the AR(1) process is
The ACF of the AR(1) process is
. For
all autocorrelations are positive, for
they
alternate between positive and negative, see Figure
11.3.
Example 12.2 (AR(2))
The AR(2) process with
,
is stationary when given the roots
and
of the
quadratic equation
it holds that
and
. We obtain solutions as
and
. Due to
and
it
holds that
and
|
(12.11) |
From the Yule-Walker equations in the case of an AR(2) process
it follows that
. The case
is excluded because a root would lie on the unit circle
(at 1 or -1). Thus for a stationary AR(2) process it must hold
that
from which, together with (
11.11), we obtain the
`stationarity triangle'
i.e., the region in which the AR(2) process is stationary.
The ACF of the AR(2) process is recursively given with
(11.12), (11.13) and
for . Figure
(11.4) displays typical patterns.