Scribd Logo
Upload

Welcome to Scribd!

  • 5 views

    Ch9 - Stochastic Trends and Seasonality

    Uploaded by

    zlxannie

    Copyright:

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

    Ch9 - Stochastic Trends and Seasonality

    Uploaded by

    zlxannie
    5 views26 pages

    Original Title

    Ch9_Stochastic Trends and Seasonality

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    5 views26 pages

    Ch9 - Stochastic Trends and Seasonality

    Uploaded by

    zlxannie

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    You are on page 1of 26

    Ch9 Stochastic Trends and Seasonality

    Forecasting – Reason and Examples

    • Overview

    • Why there is trend/seasonality?


    Given an ARMA model, we can get its characteristic equation
    as

    if the characteristic roots trend or seasonability

    • and real root: trend (constant, linear, polynomial trend)


    • complex roots & : seasonality
    • One pair of complex roots, () one seasonality
    • two pairs of complex roots, (), () two seasonalities

    2
    Stochastic Trends – Constant Trend

    • Constant trend:
    • Data has a tendency to stay at the same level.
    • AR(1). , if , ,
    • ,
    • is the constant trend operator

    • Remarks
    • The time series tends to remain the same level.
    • This level changes randomly due to the stochastic nature of
    the trend.
    • For a high order system, the constant trend occurs if one
    root 1, and other roots are smaller than one.

    3
    Stochastic Trends
    – Constant Trend Example 1
    • Example: IBM stock price

    4
    Stochastic Trends
    – Constant Trend Example 2
    • Paper making data

    5
    Stochastic Trends – Linear Trend

    • If two real roots close to 1(), linear trend


    • For an ARMA(2,1) model ,

    If
    • may linearly increase or decrease with (depends
    on ).

    6
    Stochastic trends – Linear Trend Example

    • Model
    • is the linear trend operator.
    • The data series tends to have a linear trend.
    • The slope is changing in stochastic nature due to .

    7
    Stochastic trends – Polynomial Trend and Example

    • If the Green’s function is a polynomial of j’s, i.e.:

    • For Example, An ARMA (3,2) model:

    8
    Stochastic seasonality (or periodic trends)

    • If one pair of roots are complex AND , then there is a


    stochastic seasonality.
    • Example: , and

    9
    Stochastic seasonality – Periodicity of 12
    • Periodicity of 12 (yearly) and
    Since ,
    :
    ARMA(2,1) model that gives a periodicity of 12:

    10
    Stochastic seasonality – Periodicity of 3

    • Periodicity of 3 (quarterly) and


    • ,
    • Operator
    ARMA(2,1) model

    11
    Stochastic seasonality – Arbitrary period

    • Periodicity of p (arbitrary period)


    • , ,
    operator
    ARMA(6,5) ARMA(4,3)

    12
    Table 9.1

    13
    Modeling of time series – stochastic trends/seasonalities
    Time Series Data

    ARMA(n,m) model (Ch4)

    of AR and of MA
    Remove them from the model
    Yes
    Chech if & re-estimate the lower
    No order ARMA model
    p real roots , complex roots,
    () seasonality
    use operator use operator

    Refit model

    F-test use operator

    use to model
    Seasonality adjustment or dynamic calibration

    • “Deseasonalize” data – seasonality adjustment:

    If , are a pair of complex roots, ,


    we can remove seasonality in time series presented by
    with

    or

    (Dynamic calibration of a system)

    15
    Example: Monthly average money market rates

    16
    modeling of monthly average money
    market rates

    Both ARMA (4,3) model and (1-B)(1+B2)(1,3) model are 17


    Monthly average investment data

    18
    Modeling of Monthly average investment

    19
    Only ARMA (4,3) model is adequate.
    Green’s function and its components

    20
    Example: hospital daily inpatient data

    21
    Modeling of hospital daily inpatient data

    22
    Characteristics roots of hospital data

    23
    Analysis of characteristics roots of hospital data

    24
    Example: airline ticket sales data

    25
    Analysis of Characteristic roots

    26

    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