Scribd
Upload
102 views

Chapter 1 Forecasting ARIMA Method

The Box-Jenkins method, also known as ARIMA modeling, is used to forecast univariate time series data that may exhibit trends, seasonality, cyclical patterns, or random fluctuations. It is an iterative process that identifies the underlying patterns in the data and fits an appropriate ARIMA model to minimize errors. A key assumption is that the data is stationary, which can be achieved through differencing such as subtracting the previous period's value. Computer programs facilitate applying the Box-Jenkins approach.

Uploaded by

Ajit Pal Singh
Copyright
© © All Rights Reserved
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
102 views

Chapter 1 Forecasting ARIMA Method

The Box-Jenkins method, also known as ARIMA modeling, is used to forecast univariate time series data that may exhibit trends, seasonality, cyclical patterns, or random fluctuations. It is an iterative process that identifies the underlying patterns in the data and fits an appropriate ARIMA model to minimize errors. A key assumption is that the data is stationary, which can be achieved through differencing such as subtracting the previous period's value. Computer programs facilitate applying the Box-Jenkins approach.

Uploaded by

Ajit Pal Singh
Copyright
© © All Rights Reserved
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 19

Chapter 1: Forecasting

The BoxJenkins Method


of Forecasting
Introduction
In 1970 George Box and Gwilym Jenkins
popularized ARIMA (Autoregressive
Integrated Moving Average) models in
their seminal textbook, Time Series
Analysis: Forecasting and Control.
Technically, the forecasting technique
described in the text is an ARIMA model,
however many forecasters (including the
author) use the phrases "ARIMA models"
and "Box-Jenkins models"
interchangeably.
Univariate Data
Stationary
Trend
Seasonality
Cyclical

A majority of the real-world data show


certain combinations of the above
patterns.
BoxJenkins Method

Besides the smoothing


techniques, what other
methods can we use to forecast
univariate data?
Using BoxJenkins Methods

Capture the past pattern Forecast the future


Why Use The BoxJenkins
Method?

When facing very complicated data


patterns such as a combination of a
trend, seasonal, cyclical, and random
fluctuations:
e.g. Earning data of a corporation
e.g. Forecasting stock price
e.g. Sales forecasting
e.g. Energy forecasting (electricity,
gas)
Why Use the BoxJenkins
Method?

When forecasting is the sole purpose


of the model.
Very reliable especially in short term
(06 months) prediction; reliable in
short-to-mid (6 months1.5years)
-term prediction.
Confidence intervals for the estimates
are easily constructed.
Pattern I: Stationary

Pattern 1: No TrendStationary
demand seems to cluster around a specific
level.
Pattern II: Trend

Demand
consistent
ly
increases
or Tim
e
Time

decreases
over time.

Tim Time
e
Pattern III: Seasonality
-
10.0
12.0
14.0
16.0
18.0

2.0
4.0
6.0
8.0
1 9 6 4 /1
1 9 6 5 /9
1 9 6 7 /5
1 9 69 /1
1 9 7 0 /9
1 9 7 2/5
1 9 7 4 /1
1 9 7 5 /9
1 9 7 7/5
1 9 7 9 /1
1 9 8 0/9
1 9 8 2 /5
1 9 8 4 /1
19 8 5 /9
1 9 8 7 /5
1 9 8 9/1
1 9 9 0 /9
1 9 9 2 /5
U.S. 3-Month Treasury Bill Rate

1 9 9 4 /1
Pattern IV: Cyclical

1 9 9 5 /9
1 9 9 7/5
1 9 9 9 /1
2 0 0 0 /9
2 0 0 2/5
BoxJenkins Method
Assumption

In order to use the B/J method, the time


series should be stationary.
B/J main idea: Any stationary time series
can self-predict its own future from the
past data.
10
9
8
7
6
5
4
3
2
1
0
1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85
BoxJenkins Method
Assumption

We know that not all time series are


stationary.
However, it is easy to convert a trend or
a seasonal time series to a stationary
time series.
Simply use the concept of
Differencing.
Example of Differencing
Convert a trend time series to
stationary time series using the
differencing method
0.0
2.0
4.0
6.0
8.0
10.0
12.0

-2.0
1 9 9 0 /Q 2

1 9 9 1/Q 1
1 9 9 1 /Q 4

1 9 9 2/Q 3

1 9 9 3 /Q 2

1 9 9 4/Q 1

1 9 9 4 /Q 4

1 9 9 5/Q 3

1 9 9 6 /Q 2

1 9 97 /Q 1
d t Yt Yt 4

1 9 97 /Q 4
series

1 9 98 /Q 3
U.S. Electricity Consumption

1 9 99 /Q 2

2 0 00 /Q 1

2 0 00 /Q 4

2 0 01 /Q 3
series to stationary time

2 0 02 /Q 2

2 0 03 /Q 1
Differencing Summary
To convert trend time series to
stationary time series:
d t Yt Yt 1
To convert seasonal time series to
stationary time series:
d t Yt Yt 4
Both of the above two methods can
be applied/combined to remove the
cyclical effects.
BoxJenkins models capture a wide variety of
time series patterns.
When faced with a complicated time series
that includes a combination of trend, seasonal
factor, cyclical, as well as random fluctuations,
use of the BoxJenkins is appropriate.
This methodology for forecasting is an
iterative process that begins by assuming a
tentative pattern that is fitted to the data so
that error will be minimized.
The major assumption of the model is that the
data is stationary.
Differencing could be used to make the data
stationary.
Computer programs such as Minitab, and SPSS
provide all the analysis tools for using the Box
Jenkins.

You might also like