Chapter 2

Data Patterns and Choice

of Forecasting Techniques
Lectured by Mr. Oussa em

Business Forecasting
Chapter Topics

Data Patterns

Forecasting Methodologies

Technique Selection

Model Evaluation
Data Pattern and Choice of Technique

• The pattern of data

• The nature of the past relationship in the data

• The level of subjectivity in making a forecast

Data Pattern and Choice of Technique

Multivariate
Univariate forecasting Qualitative forecasts
forecasting techniques
techniques depend on: depend on:
depend on

Subjectivity:
Past data Past
Forecasters
patterns. relationships.
intuition.
Data Patterns

Data Patterns as a Guide

•Simple observation of the data will show the way

that data have behaved over time.
•Data pattern may suggest the existence of a
relationship between two or more variables.
•Four Patterns: Horizontal, Trend, Seasonal, Cyclical.
Data Patterns
• Horizontal
• When there is no trend in the data pattern, we deal with horizontal data pattern.

Forecast Variable

Mean

Time
Data Patterns
• Trend
• Long-term growth movement of a time series

Trend Yt Trend
Yt

t t

Yt Trend Yt
Trend

t t
Data Patterns
• Seasonal Pattern
• A predictable and repetitive movement observed around a trend line within a
period of 1 year or less.

Forecast Variable

Time
Data Patterns
• Cyclical

• Occurs with business and economic expansions and contractions.

• Lasts longer than 1 year.

• Correlated with business cycles.

Other Data Patterns
• Autocorrelated Pattern
• Data in one period are related to their values in the previous period.

• Generally, if there is a high positive autocorrelation, the value in the

month of June, for example, is positively related to the values in the
month of May.

• This pattern is more fully discussed when we talk about the Box–
Jenkins methodology.
 Measures of Accuracy in Forecasting
• Error in Forecasting

et = Yt − Yˆt
• Measures the average error that can be expected over time.
• The average error concept has some problems with it. The positive and
negative values cancel each other out and the mean is very likely to be close
to zero.
 Error in Forecasting
• Mean Average Deviation (MAD)

e t
MAD = t =1
n
 Error in Forecasting
• Mean Square Error (MSE)


t =1
2
(e t )
MSE =
n
 Error in Forecasting
• Mean Absolute Percentage Error

n
(et / Yt ) 100
MAPE = 
t =1
n
 Error in Forecasting
• Mean Percentage Error

 (e
t =1
t / Yt )
MPE =
• No bias, MPE should be zero. n
 Evaluating Reliability

• Forecasters use the following two approaches to determine if the

forecast is reliable or not:
• Root Mean Square (RMS)


t =1
et2
RMS =
n
 Evaluating Reliability
• Root Percent Mean Square (R%MS)

t =1
(et2 / Yt )
R % MS =
n
 Forecasting Methodologies

• Forecasting methodologies fall into three categories:

• Quantitative Models

• Qualitative Models

• Technological Approaches
 Forecasting Methodologies
• Quantitative Models
• Also known as statistical models.

• Include time series and regression approaches.

• Forecast future values entirely on the historical observation of a

variable.
 Forecasting Methodologies
• Quantitative Models
• An example of a quantitative model is shown below:

Yt +1 =  0 + 1Yt +  2Yt −1
Yt +1 = Sales one time period into the future
Yt = Sales in the current period
Yt −1 = Sales in the last period
 Forecasting Methodologies
• Qualitative Models
• Non-statistical or judgment models

• Expert opinion

• Executive opinion

• Sale force composite forecast

• Focus groups

• Delphi method
 Forecasting Methodologies
• Technological Approach
• Combines quantitative and qualitative methods.

• The objective of the model is to combine technological, societal,

political, and economic changes.
 Technique Selection
• Forecasters depend on:
• The characteristics of the decision making situation which
may include:
• Time horizon

• Planning vs. control

• Level of detail

• Economic conditions in the market (stability vs. state of flux)

 Technique Selection
• Forecasters depend on:
• The characteristics of the forecasting method
• Forecast horizon

• Pattern of data

• Type of model

• Costs associated with the model

• Level of accuracy and ease of application

 Model Evaluation
• Forecasters depend on:
• The level of error associated with each model.

• Error is computed and looked at graphically.

• Control charts are used for model evaluations.

• Turning point diagram is used to evaluate a model.

 Model Evaluation
• A pattern of cumulative errors moving systematically away from zero
in either direction is a signal that the model is generating biased
forecasts.
• Management has to establish the upper and lower control limits.
• One fairly common rule of thumb is that the control limits are equal
to 2 or 3 time the standard error.
 Model Evaluation

30
25
20
15
Cumulative Error

10 Model A
5 Model B
0 Model C
-5 Model D
-10
-15
-20
-25
Time
 Model Evaluation
Actual Change
Y

Line of Perfect Forecast

II–Turning Point Error IB–Underestimate
Prediction of downturn of Positive
that did not occur; or change
failure to predict an IA–Overestimate
upturn of Positive change

IIIA–Overestimate of Yˆ Forecast

Negative change Change
IV–Turning Point Error
Prediction or upturn
that did not occur; or failure
to predict a downturn
IIIB–Underestimate
of Negative change

Figure 2.6 Turning Point Error Diagram

 Model Evaluation
Actual Change
400

300

200
Forecast Change

100

0
-300 -200 -100 0 100 200 300 400
-100

-200

-300

-400

Figure 2.7 Turning Point Analysis for Model C

 Data Patterns

Forecasting Methodologies
Chapter
Summary Technique Selection

Model Evaluation