10 - Notes 10 Seasonal Models : Let s denote the known...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Notes 10 Seasonal Models : Let s denote the known seasonal period (eg. for monthly series, s = 12). Define Seasonal MA( Q ) s model by Z t = a t- Θ 1 a t- s- Θ 2 a t- 2 s- ...- Θ Q a t- Qs Seasonal MA(1) 12 Z t = a t- Θ a t- 12 , | Θ | < 1 Cov ( Z t ,Z t- 1 ) = Cov ( a t- Θ 1 a t- 12 ,a t- 1- Θ a t- 13 ) = 0 . Cov ( Z t ,Z t- 12 ) = Cov ( a t- Θ 1 a t- 12 ,a t- 12- Θ a t- 24 ) =- Θ σ 2 a . ρ k = 1 , k = 0 ,- Θ 1+Θ 2 , | k | = 12 , , | k | 6 = 0 , 12 Rewrite Seasonal MA( Q ) s model as Z t = Θ( B ) a t where Θ( x ) = 1- Θ 1 x s- Θ 2 x 2 s- ...- Θ Q x Qs is the seasonal MA characteristic polynomial. Also we have ρ = 1 ρ ks =-- Θ k + Θ 1 Θ k +1 + Θ 2 Θ k +2 + ... + Θ Q- k Θ Q 1 + Θ 2 1 + Θ 2 2 + ... + Θ 2 Q , k = 1 , 2 ,...,Q. and ρ h = 0 for h is not a multiple of s of h > Qs . Note that seasonal MA( Q ) s is always stationary and invertible if and only if all roots of Θ( x ) = 0 all exceed one in absolute value. Also, the model can also be viewed as an MA( Qs ) model with all θ ’s equals to zero except at the seasonal lags s, 2 s,...,Qs ....
View Full Document

This note was uploaded on 11/02/2011 for the course STAT 4005 taught by Professor Wu,kaho during the Spring '08 term at CUHK.

Page1 / 4

10 - Notes 10 Seasonal Models : Let s denote the known...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online