VIEWS: 184 PAGES: 30

• pg 1
```									  Business Forecasting

Chapter 2
Data Patterns and Choice of
Forecasting Techniques
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

All of the above help us in how we classify the
forecasting technique.
Data Pattern and Choice of
Technique

   Univariate forecasting techniques depend on:
   Past data patterns.

   Multivariate forecasting techniques depend on
   Past relationships.

   Qualitative forecasts depend on:
   Subjectivity: Forecasters 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.

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
Measures of Accuracy in Forecasting

       Error in Forecasting
ˆ
et  Yt  Yt
   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

n

e        t
n
Error in Forecasting

   Mean Square Error (MSE)

n

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
n

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

   Forecasters use the following two approaches
to determine if the forecast is reliable or not:
   Root Mean Square (RMS)

n


t 1
e t2
RMS 
n
Evaluating Reliability

   Root Percent Mean Square (R%MS)

n


t 1
(e t2 / 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                                          ˆ
Y 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
Chapter Summary

   Data Patterns

   Forecasting Methodologies

   Technique Selection

   Model Evaluation

```
To top