Econ 512 Box Jenkins Slides

Download as pdf or txt
Download as pdf or txt
You are on page 1of 31
 
Box-Jenkins Analysis of ARMA(p,q) Models
Eric ZivotApril 7, 2011
 
Box-Jenkins Modeling Strategy for Fitting ARMA(
 p,q
)
 Models1. Transform the data, if necessary, so that the assumption of covariancestationarity is a reasonable one2. Make an initial guess for the values of 
 p
 and
 q
3. Estimate the parameters of the proposed ARMA(
 p,q
)
 model4. Perform diagnostic analysis to con
rm that the proposed model adequatelydescribes the data (e.g. examine residuals from
 
tted model)
5
 
Identi
cation of Stationary ARMA(p,q) ProcessesIntuition: The mean, variance, and autocorrelations de
ne the properties of an ARMA(p,q) model. A natural way to identify an ARMA model is to matchthe pattern of the observed (sample) autocorrelations with the patterns of thetheoretical autocorrelations of a particular ARMA(
 p,q
)
 model.Sample autocovariances/autocorrelations
ˆ
γ 
 j
 = 1
X
t
=
 j
+1
(
y
t
ˆ
μ
)(
y
t
 j
ˆ
μ
)
,
 ˆ
μ
 = 1
X
t
=1
y
t
ˆ
ρ
 j
 =ˆ
γ 
 j
ˆ
γ 
0
Sample autocorrelation function (SACF)/correlogramplot
 ˆ
ρ
 j
 vs.
 j
5
 
Result: If 
 
t
W
(0
,σ
2
)
 then
 ρ
 j
 = 0
 for all
 j
 and
√ 
ˆ
ρ
 jd
(0
,
1)
so that
avar
ρ
 j
) = 1
Therefore, a simple
 t
statistic for
 
0
 :
 ρ
 j
 = 0
 is
√ 
ˆ
ρ
 j
and we reject
 
0
 :
 ρ
 j
 = 0
 if 
|
ˆ
ρ
 j
|
 >
 1
.
96
√ 
Remark:
 ˆ
ρ
1
,...,
ˆ
ρ
k
 are asymptotically independent:
√ 
ˆ
ρ
kd
(
0
,
I
k
)
where
 ˆ
ρ
k
 =
ρ
1
,...,
ˆ
ρ
k
)
0
.
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5

Reward Your Curiosity

Everything you want to read.
Anytime. Anywhere. Any device.
No Commitment. Cancel anytime.
576648e32a3d8b82ca71961b7a986505