Ex2_2005

# Ex2_2005 - Y t = p X j =1 φ j,p Y t-j ± t Show that the...

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

Time Series (MA5/30085) 2005 Exercises 2 In these questions, { ± t } is a discrete, purely random process, such that E ( ± t ) = 0, V AR ( ± t ) = σ 2 ± , COV ( ± t t + τ ) = 0 for τ 6 = 0. 1. Find the ACF of the second order MA process given by X t = ± t + 0 . 7 ± t - 1 - 0 . 2 ± t - 2 2. Show that the ACF of he m th order MA process is given by X t = m k =0 ± t - k / ( m + 1) is ρ τ = ( m + 1 - k ) / ( m + 1) k = 0 , 1 ,...,m 0 k > m 3. Show that the inﬁnite MA process { ± } deﬁned by X t = ± t + C ( ± t - 1 + ± t - 2 + ... ) where C is a constant is non-stationary. Also show that the series of ﬁrst diﬀerences { Y t } deﬁned by Y t = X t - X t - 1 is a ﬁrst order MA process and is stationary. Find the ACF of { Y t } 4. The stationary process { X t } has ACVF s τ . A new stationary process { Y t } is deﬁned as Y t = X t - X t - 1 . Obtain the ACVF of { Y t } in terms of s τ and ﬁnd form when s τ = λ | k | ( λ is a constant). 5. (a) What is meant by saying that a linear process is invertible? (b) How may we check whether such a process is invertible? (c) Determine whether the following process is invertible: X t = 4 25 ± t + 4 25 ± t - 1 + 1 25 ± t - 2 . 6. Suppose that { Y t } is a zero mean stationary AR( p ) process:

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

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Y t = p X j =1 φ j,p Y t-j + ± t . Show that the process { X t } given by X t = q X k =0 β k Y t-k , where β = 1, can be written as an ARMA( p , q ) process. 1 7. HARDER QUESTION (a) Let { X t } be a real zero mean stationary process with variance σ 2 X and let X = 1 N ∑ N t =1 X t . Show that we can write var { X } = σ 2 X N 1 + 2 N N-1 X i =1 N X j>i ρ j-i . where { ρ τ } is the autocorrelation sequence of { X t } . (b) Now consider the AR(1) process X t = φ 1 , 1 X t-1 + ± t with variance σ 2 X = σ 2 ± / (1-φ 2 1 , 1 ) and autocorrelation sequence ρ τ = φ | τ | 1 , 1 . Show that for this process N X j>i ρ j-i = φ 1 , 1 (1-φ N-i 1 , 1 ) 1-φ 1 , 1 . (c) Hence show that for the AR(1) process var { X } = σ 2 ± N (1-φ 2 1 , 1 ) ± 1 + 2 φ 1 , 1 N (1-φ 1 , 1 ) " N-(1-φ N 1 , 1 ) (1-φ 1 , 1 ) #! . 2...
View Full Document

## This note was uploaded on 10/19/2009 for the course MATH 611 taught by Professor Jsdkasj during the Spring '09 term at Kansas.

### Page1 / 2

Ex2_2005 - Y t = p X j =1 φ j,p Y t-j ± t Show that the...

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

View Full Document
Ask a homework question - tutors are online