Scribd
Upload
7 views7 pages

ARIMA and VARIMA Forecasting Methods

The document discusses various time series forecasting methods including: 1) Simple classical methods like naive models, moving averages, exponential smoothing, decomposition, and regression; 2) ARIMA models using the Box-Jenkins procedure; 3) Double seasonal ARIMA models; 4) Intervention and outlier detection models; 5) Calendar variation and transfer function models; 6) VARIMA and GSTAR models; and 7) Introduction to nonlinear time series models. It also discusses characteristics of forecasting methods including objective versus subjective approaches, and patterns in time series like trends and seasonality.

Uploaded by

Amira Rana
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
7 views7 pages

ARIMA and VARIMA Forecasting Methods

The document discusses various time series forecasting methods including: 1) Simple classical methods like naive models, moving averages, exponential smoothing, decomposition, and regression; 2) ARIMA models using the Box-Jenkins procedure; 3) Double seasonal ARIMA models; 4) Intervention and outlier detection models; 5) Calendar variation and transfer function models; 6) VARIMA and GSTAR models; and 7) Introduction to nonlinear time series models. It also discusses characteristics of forecasting methods including objective versus subjective approaches, and patterns in time series like trends and seasonality.

Uploaded by

Amira Rana
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd

Material

1. Introduction
[Link] Classical Forecasting Methods:
Naïve Models, Moving Average Methods,
Exponential Smoothing Methods,
Decomposition Method, Regression and
Trend Analysis, and Time Series
Regression.
[Link] models (Box-Jenkins procedure)
[Link] Seasonal ARIMA
[Link] Model and Outlier Detection
[Link] Variation Model
[Link] Function Model
[Link] and GSTAR Model
[Link] to Nonlinear Time Series
1
Model Module
and 1: S1Modern Methods
Statistics, ITS, February 11, 2013for Time
Series Forecasting
Characteristics of Forecasting
Methods
Source: J.S. Armstrong, “Principles of
Forecasting”

2 Module 1: S1 Statistics, ITS, February 11, 2013


Intro: Forecasting Methods
Forecasting Method

Objective Subjective (Judgmental)


Forecasting Methods Forecasting Methods

Time Series Causal


Analogies
Methods Methods

Naïve Methods Simple Regression


Delphi
Moving Averages Multiple Regression
PERT
Exponential Smoothing Neural Networks

Simple Regression
Survey techniques
ARIMA

Neural Networks
References :
 Makridakis et al. (1998)
Combination of Time Series – Causal Methods
 Hanke and Reitsch (2004)
 Intervention Model
 Wei, W.W.S.
 Transfer Function (ARIMAX)
(2006) 
 VARIMA (VARIMAX)
Box, Jenkins and Reinsel (1994)
 Neural Networks

3 Module 1: S1 Statistics, ITS, February 11, 2013


Time Series
PATTERN
General
time series
“PATTERN”
Stationer
 Trend:
linear & nonlinear

Seasonal: additive &
4 multiplicative
Module 1: S1 Statistics, ITS, February 11, 2013
General of Time Series
PATTERN

5 Module 1: S1 Statistics, ITS, February 11, 2013


Motivation
 The M3-Competition: results,
conclusions and implications
(a) Statistically sophisticated or complex
methods do not necessarily provide
more accurate forecasts than simpler
ones.
(b) The relative ranking of the performance of
the various methods varies according to the
accuracy measure being used.
(c) The accuracy when various methods are
being combined outperforms, on average,
the individual methods being combined and
does very well in comparison to other
methods.
6
(d) The accuracy of the
Module 1: S1 Statistics, ITS, various methods
February 11, 2013
depends upon the length of the forecasting
Selecting Forecasting Methods

Source: J.S. Armstrong, “Principles of


Forecasting”

7 Module 1: S1 Statistics, ITS, February 11, 2013

You might also like