Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • Exercise Set 3: Learning Outcomes: Working With AR (P) and MA (Q) Processes and Their Characteristics

    Uploaded by

    Harshvardhan Singh
    11 views1 page
    1. This document contains 5 exercises on analyzing the properties of AR(p) and MA(q) time series models. It covers computing the mean, autocovariance and autocorrelation functions of AR(1) processes, bounds on the autocorrelation of MA(1) processes, the variance and autocorrelation structure of MA(q) processes, and using recursive relations and Yule-Walker equations to analyze AR(p) models.

    Original Description:

    Original Title

    ex3

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    1. This document contains 5 exercises on analyzing the properties of AR(p) and MA(q) time series models. It covers computing the mean, autocovariance and autocorrelation functions of AR(1) processes, bounds on the autocorrelation of MA(1) processes, the variance and autocorrelation structure of MA(q) processes, and using recursive relations and Yule-Walker equations to analyze AR(p) models.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    Download as pdf or txt
    11 views1 page

    Exercise Set 3: Learning Outcomes: Working With AR (P) and MA (Q) Processes and Their Characteristics

    Uploaded by

    Harshvardhan Singh
    1. This document contains 5 exercises on analyzing the properties of AR(p) and MA(q) time series models. It covers computing the mean, autocovariance and autocorrelation functions of AR(1) processes, bounds on the autocorrelation of MA(1) processes, the variance and autocorrelation structure of MA(q) processes, and using recursive relations and Yule-Walker equations to analyze AR(p) models.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Flag for inappropriate content
    Download as pdf or txt
    of 1

    FM05 – Exercise Set 3

    1. Consider the stationary solution of an AR(1) model:


    ∞
    X
    Xt = φi1 εt−i ,
    i=0

    with |φ1 | < 1 and εt a white noise process with variance σε2 . For Xt compute:
    a. the mean function,
    b. the autocovariance function,
    c. the autocorrelation function.
    2. a. Show that for an M A(1) process Xt = εt + θ1 εt−1 we have:
    1
    |ρ(1)| ≤ , ∀θ1 ∈ R.
    2
    b. For what values of θ1 does ρ(1) attain its maximum and minimum?
    3. a. For an M A(q) process show that the variance is:

    γ(0) = (1 + θ12 + · · · + θq2 )σε2 ,

    and that the autocorrelation function has non-zero spikes:


    θh + θ1 θh+1 + θ2 θh+2 + · · · + θq−h θq
    ρ(h) = , h = 1, . . . , q.
    1 + θ12 + · · · + θq2

    b. Compute and plot the autocorrelation function ρ(h), h = 0, . . . , 5 for an M A(2) process with:
    i. θ1 = 0.5, θ2 = 0.4,
    ii. θ1 = 1.2, θ2 = −0.7,
    iii. θ1 = −1, θ2 = −0.6.
    4. Consider an AR(2) process of the form Yt = φ1 Yt−1 + φ2 Yt−2 + εt . Use the recursive relation:

    ρ(h) = φ1 ρ(h − 1) + φ2 ρ(h − 2)

    to calculate and plot the first 6 values of the autocorrelation function ρ(h) when:
    a. φ1 = 0.6, φ2 = 0.3,
    b. φ1 = −0.4, φ2 = 0.5,
    c. φ1 = −1, φ2 = −0.6.
    5. Consider the stationary AR(3) process:
    1 7 1
    Yt = Yt−1 + Yt−2 − Yt−3 + εt .
    6 18 9

    a. Write down the Yule-Walker equations for this process.


    b. Solve the Yule-Walker equations to find values for ρ(1), ρ(2) and ρ(3).
    c. Calculate and plot the autocorrelation function for this process up to lag h = 5.

    Learning outcomes: working with AR(p) and MA(q) processes and their characteristics.

    You might also like