Scribd Logo
Upload

Welcome to Scribd!

  • 21 views

    1 Time Series and Stationarity: Example Class 8

    Uploaded by

    Marvin Lie
    1. The document provides an overview of concepts related to time series analysis including stationarity, auto-covariance function, auto-correlation function, and examples of stationary and non-stationary time series. 2. It examines examples including a moving average process to determine if it is weakly stationary and examines the autocorrelation function of a signal plus noise time series. 3. The final example determines if a time series consisting of a cosine and sine function with random coefficients is weakly stationary.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    1 Time Series and Stationarity: Example Class 8

    Uploaded by

    Marvin Lie
    21 views2 pages
    1. The document provides an overview of concepts related to time series analysis including stationarity, auto-covariance function, auto-correlation function, and examples of stationary and non-stationary time series. 2. It examines examples including a moving average process to determine if it is weakly stationary and examines the autocorrelation function of a signal plus noise time series. 3. The final example determines if a time series consisting of a cosine and sine function with random coefficients is weakly stationary.

    Original Title

    T8 (3)

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    1. The document provides an overview of concepts related to time series analysis including stationarity, auto-covariance function, auto-correlation function, and examples of stationary and non-stationary time series. 2. It examines examples including a moving average process to determine if it is weakly stationary and examines the autocorrelation function of a signal plus noise time series. 3. The final example determines if a time series consisting of a cosine and sine function with random coefficients is weakly stationary.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    21 views2 pages

    1 Time Series and Stationarity: Example Class 8

    Uploaded by

    Marvin Lie
    1. The document provides an overview of concepts related to time series analysis including stationarity, auto-covariance function, auto-correlation function, and examples of stationary and non-stationary time series. 2. It examines examples including a moving average process to determine if it is weakly stationary and examines the autocorrelation function of a signal plus noise time series. 3. The final example determines if a time series consisting of a cosine and sine function with random coefficients is weakly stationary.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 2

    21/22

    THE UNIVERSITY OF HONG KONG


    DEPARTMENT OF STATISTICS AND ACTUARIAL SCIENCE
    STAT3907 Linear Models and Forecasting (2021-2022)

    EXAMPLE CLASS 8

    1 Time Series and Stationarity


    1. Time series data: all observed time series data (z1 , z2 , ..., zT ) is a realization of T
    random variables (Z1 , Z2 , ..., ZT ).

    2. Time series: a time series is a sequence of random variables, which are ordered by
    time, and denoted by {Zt , t ∈ Z}.

    3. Auto-covariance function (ACVF):

    Cov(Zt , Zs ) = E[(Zt − µt )(Zs − µs )]


    = E(Zt Zs ) − µt µs

    A useful formula
    " m n
    # m X
    n
    X X X
    Cov ci Y i , dj Zj = ci dj Cov(Yi , Zj ).
    i=1 j=1 i=1 j=1

    4. Auto-correlation function (ACF):

    Cov(Zt , Zs ) γ(t, s)
    ρ(t, s) = Corr(Zt , Zs ) = p =p
    V ar(Zt )V ar(Zs ) γ(t, t)γ(s, s)

    5. Strict stationarity: a time series {Zt } is said to be strictly stationary if the joint
    distribution of Zt1 , Zt2 , . . . , Ztn is the same as that of Zt1 −k , Zt2 −k , . . . , Ztn −k for
    all choices of natural number n, all choices of time points t1 , t2 , . . . , tn and all
    choices of time lag k.

    6. Weak stationarity: A time series {Zt } is said to be weakly (second-order, or


    covariance) stationary if

    (a) the mean function µt is constant over time, and


    (b) γ(t, t − k) = γ(0, k) for all times t and lags k.

    7. White noise: A stationary process is said to be a white noise process if it is a


    sequence of i.i.d random variables {at } with zero mean and variance σa2 > 0. Such
    a white noise is usually abbreviated as W N (0, σa2 ).

    1
    2 Examples
    1. Suppose that {Zt , t ∈ Z} is given by

    Zt = at − 0.5at−1 , t ∈ Z, (1)

    where the a’s are assumed to be a sequence of i.i.d. random variables with zero
    mean and variance σa2 .

    (a) The mean function of {Zt };


    (b) The variance function of {Zt };
    (c) The auto-covariance function (ACVF);
    (d) The auto-correlation function (ACF).

    2. Revisit the moving average example, Zt = at − 0.5at−1 in Question 1, where the


    at ’s are assumed to be a sequence of i.i.d. random variables with zero mean and
    variance σa2 . Is it weakly stationary?

    3. Signal plus noise. Let {Xt } be a time series in which we are interested. However,
    because the measurement process itself is subject to error in some circumstances,
    we actually observe Zt = Xt + at . We assume that {Xt } and {at } are independent
    processes and that {at } is white noise. We call {Xt } the signal and {at } the
    measurement noise or error.
    If {Xt } is stationary with autocorrelation function ρk , show that {Zt } is also sta-
    tionary with
    ρk
    corr(Zt , Zt−k ) = 2
    , k ≥ 1,
    1 + σa2 /σX

    where σa2 = var(at ) and σX


    2
    = var(Xt ). We call σX 2
    /σa2 the signal-to-noise ratio,
    or SNR. Show that the larger the value of the SNR, the closer the autocorrelation
    function of the observed series Zt is to that of the desired signal {Xt }.

    4. Is the following time series {Xt } weakly stationary? Specify the reasons for your
    answers. Xt = A cos(tλ)+B sin(tλ), where A and B are given uncorrelated random
    variables with mean zero and variance σ 2 and λ is a given number.

    You might also like

    pFad - Phonifier reborn

    Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

    Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


    Alternative Proxies:

    Alternative Proxy

    pFad Proxy

    pFad v3 Proxy

    pFad v4 Proxy