ARMA
ARMA
ARMA
STATIONARY TS MODELS
MA(1): Xt = Zt + θZt−1
Here, as in the MA and AR models, we can use the backshift operator to write the
ARMA model more concisely as
1
θ(B) = (1 + φB + φ2 B 2 + φ3 B 3 + . . .)(1 + θB)
φ(B)
= 1 + φB + φ2 B 2 + φ3 B 3 + . . . + θB + φθB 2 + φ2 θB 3 + φ3 θB 4 + . . .
= 1 + (φ + θ)B + (φ2 + φθ)B 2 + (φ3 + φ2 θ)B 3 + . . .
= 1 + (φ + θ)B + (φ + θ)φB 2 + (φ + θ)φ2 B 3 + . . .
X∞
= ψj B j ,
j=0
Now, suppose that |φ| > 1. Then, by similar arguments as in the AR(1) model, it
can be shown that
∞
X
Xt = −θφ Zt − (φ + θ)
−1
φ−j−1 Zt+j .
j=1
Here too, we obtained a noncausal process which depends on future noise values,
hence of no practical value.
While causality means that the process {Xt } is expressible in terms of past values
of {Zt }, the dual property of invertibility means that the process {Zt } is express-
ible in the past values of {Xt }. Is ARMA(1,1) invertible?
ARMA(1,1) model is
φ(B)Xt = θ(B)Zt
and so writing the solution for Zt we have
1 1
Zt = φ(B)Xt = (1 − φB)Xt . (4.34)
θ(B) 1 + θB
86 CHAPTER 4. STATIONARY TS MODELS
1
Now, if |θ| < 1 then the power series expansion of the function f (x) = 1+θx
is
∞
1 X
f (x) = = (−θ)j xj ,
1 + θx j=0
The two properties, causality and invertibility, determine the admissible region for
the values of parameters φ and θ, which is the square
−1 <φ <1
− 1 < θ < 1.
where Zt is a white noise, is very helpful in deriving the ACVF and ACF of the
process. By Corollary 4.1 we have
∞
X
2
γ(τ ) = σ ψj ψj+τ .
j=0
(φ + θ)2
2
=σ 1+ .
1 − φ2
and
∞
X
2
γ(1) = σ ψj ψj+1
j=0
= σ 2 (φ + θ) + (φ + θ)2 φ(1 + φ2 + φ4 + . . .)
" ∞
#
X
= σ 2 (φ + θ) + (φ + θ)2 φ φ2j
j=0
(φ + θ)2 φ
2
= σ (φ + θ) +
1 − φ2
θ MA(1)
1
AR(1)
-1 1 φ
-1
φ = −θ
WN
30 80 130 180
-5
-10
phi = -0.9, theta = -0.5 phi = 0.9, theta = -0.5
10
-5
-10
30 80 130 180
1.0
0.8
0.5
0.6
ACF
ACF
0.4
0.0
0.2
-0.5
0.0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Lag Lag
1.0
0.8
0.5
0.6
ACF
ACF
0.4
0.0
0.2
-0.5
0.0
-0.2
-1.0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Lag Lag
Figure 4.16: ACF of the ARMA(1,1) processes with the parameter values as in
Figure 4.15, respectively.