2/17/2020

Forecasting

Learning Objectives
 List the elements of a good forecast.

 Outline the steps in the forecasting process.

 Compare qualitative and quantitative approaches to

forecasting.

 Quantitative- Naïve Forecasts, Averaging Forecasts,

Exponential Smoothing, Linear Trend Forecasts

1
 2/17/2020

Forecasting
Process of predicting a future event, scientifically &
with base

Taking historical data & projecting them into the

future
I see that you will
get an A this semester.

Forecasting Time Horizons

Short-range forecast
 Up to 1 year, generally less than 3 months
 Purchasing, job scheduling, workforce levels, job assignments, production
levels

Medium-range forecast
 3 months to 3 years
 Sales and production planning, budgeting

Long-range forecast
 3+ years
 New product planning, facility location, research and development

2
 2/17/2020

Types of Forecasts
Economic forecasts
 Address business cycle – inflation rate, money supply, housing
starts, etc.

Technological forecasts
 Predict rate of technological progress
 Impacts development of new products

Demand forecasts
 Predict sales of existing product

Strategic Importance of
Forecasting
 Human Resources – Hiring, training, laying off
workers

 Capacity – Capacity shortages can result in

undependable delivery, loss of customers, loss of
market share

 Supply-Chain Management – Good supplier relations

and price advance

3
 2/17/2020

Forecasting Approaches

Quantitative Methods (Calculation)

Used when situation is ‘stable’ and historical

data exist
 Existing
products
 Current technology

Involves mathematical techniques

 e.g., forecasting sales of color televisions

Forecasting Approaches

Qualitative Methods

Used when situation is vague and little data exist

 New products
 New technology

Involves intuition, experience

 Panel of experts, queried iteratively

4
 2/17/2020

Overview of Quantitative
Approaches

1. Naive approach
2. Moving averages Time-Series
Models
3. Exponential
smoothing

Associative
4. Linear Trend Model
5. Linear regression

Time Series Forecasting

Set of evenly spaced numerical data

Obtained by observing response variable at
regular time periods

Forecast based only on past values

Assumes that factors influencing past and
present will continue influence in future

10

5
 2/17/2020

11

Trend Component

 Persistent, overallupward or downward pattern

 Changes due to population, technology, age, culture, etc.
 Typically several years duration

12

6
 2/17/2020

Seasonal Component

 Regularpattern of up and down fluctuations

 Due to weather, customs, etc.
 Occurs within a single year

13

Cyclical Component
 Repeating up and down movements
 Affected by business cycle, political, and economic factors
 Multiple years duration
 Often causal or
associative
relationships

0 5 10 15 20

14

7
 2/17/2020

Random Component
 Erratic, unsystematic, ‘residual’
fluctuations
 Due to random variation or unforeseen events
 Short duration and
non repeating

M T W T F

15

1. NAIVE APPROACH

 Assumes demand in next period is the

same as demand in most recent period
 e.g., If May sales were 48, then June sales will be 48
 Sometimes cost effective and efficient
 Cannot provide high accuracy
 Quick and easy to prepare

16

8
 2/17/2020

17

TECHNIQUES FOR AVERAGING

 Moving average
 Weighted moving average
 Exponential smoothing

18

9
 2/17/2020

2. Moving Average Method (MA)

A technique that averages a number of recent actual

values, updated as new values become available.

 Used if little or no trend

∑ demand in previous n periods

Moving average = n

19

3 Month Moving Average Example

Actual 3-Month
Month Shed Sales Moving Average
January 10
February 12
March 13
April 16 (10 + 12 + 13)/3 = 11.67 ~12
May 19 (12 + 13 + 16)/3 = 13.67~14
June 23 (13 + 16 + 19)/3 = 16
July 26 (16 + 19 + 23)/3 = 19.33~20

ROUND UP
20

10
 2/17/2020

3. Weighted Moving Average (WMA)

 Used when trend is present

 Weights based on experience and intuition
 More recent values in a series are given more weight in
computing the forecast

∑ (weight for period n)

Weighted x (demand in period n)
moving average = ∑ weights

21

Weighted Moving Average

Example Weights Applied Period
3 Last month
2 Two months ago
1 Three months ago
6 Sum of weights

Actual 3-Month Weighted

Month Shed Sales Moving Average
January 10
February 12
March 13
April 16 [(3 x 13) + (2 x 12) + (1x10)]/6 = 12.16~13
May 19 [(3 x 16) + (2 x 13) + (12)]/6 = 14.33~15
June 23 [(3 x 19) + (2 x 16) + (13)]/6 = 17
July 26 [(3 x 23) + (2 x 19) + (16)]/6 = 20.5~21

ROUND UP
22

11
 2/17/2020

Potential Problems With

Moving Average

 Increasing n smooths the forecast but makes it less

sensitive to changes

 Do not forecast trends well

 Require extensive historical data

23

Common Measures of Error

1. Mean Absolute Deviation (MAD)

∑ |actual - forecast|
MAD =
ne
2. Mean Squared Error (MSE)

MSE = ∑ (|actual - forecast|)2

ne-1
3. Mean Absolute Percentage Error (MAPE)
( |actual - forecast| / actual*100)
MAPE =
ne
24

12
 2/17/2020

MAD, MSE & MAPE Example

Period Actual Forecast A-F |A-F| |A-F|^2 (|A-F|/Actual)*100

1 217 215 2
2 213 216 -3
3 216 215 1
4 210 214 -4
5 213 211 2
6 219 214 5
7 216 217 -1
8 212 216 -4

25

Tutorial 1

26

13
 2/17/2020

Actual WMA Naïve

Sep 230
Oct 304
Nov 415
Dec 420

|AD|
Sep
Oct
Nov
Dec
Sum
MAD

27

4. Exponential Smoothing

 Form of weighted moving average

 Weights decline exponentially
 Most recent data weighted most

 Requires smoothing constant ()

 Ranges from 0 to 1
 Subjectively chosen

 Involves little record keeping of past data

28

14
 2/17/2020

Formula Provided

Ft = Ft-1 + (At-1-Ft-1)

where Ft = new forecast

Ft – 1 = previous forecast

At-1 = previous actual demand

 = smoothing (or weighting)

constant (0  a  1)

29

Choosing 

• The objective is to obtain the most accurate forecast

no matter the technique

• We generally do this by selecting the model that

gives us the lowest forecast error

Forecast error = Actual demand – Forecast value

= At - Ft

30

15
 2/17/2020

Exponential Smoothing Example

Port of Baltimore has unloaded large quantities of grains
from ships during the past eight quarters. The port’s
operation manager wants to test he use of exponential
smoothing to see how well the technique works in
predicting tonnage unloaded. He assumes that the
forecast in the first quarter was 175 tons. Two value of
 for 0.1 and 0.5 are examined.

31

Information given for past 8 Quarters

Quarter Actual Tonnage

Unloaded
1 180
2 168
3 159
4 175
5 190
6 205
7 180
8 182

32

16
 2/17/2020

Calculate Forecast using  = 0.1

Q At Ft = Ft-1 +  (At-1-Ft-1) AD=|At-Ft |

1 180
2 168

3 159
4 175
5 190
6 205
7 180
8 182

33

Calculate Forecast using  = 0.5

Q At Ft = Ft-1 +  (At-1-Ft-1) AD=|At-Ft |

1 180
2 168
3 159
4 175

5 190
6 205
7 180
8 182

34

17
 2/17/2020

Mean Absolute Deviation (MAD)

∑ |actual - forecast|
MAD =
n
Which one is more accurate? More preferable?

MAD0.1 = =

MAD0.5 = =

35

Tutorial 2 Sep 2013

Assume 3MA

36

18
 2/17/2020

Actual Naïve 3MA α=0.3

1 370
2 310
3 450
4 290
5

Actual-Forecast=|AD|
1
2
3
4
Sum
MAD

37

5. Linear Trend Equation

Ft

Ft = a + bt

0 1 2 3 4 5 t
 Ft = Forecast for period t
 t = Specified number of time periods
 a = Value of Ft at t = 0
 b = Slope of the line

38

19
 2/17/2020

Calculating a and b; Formula Provided

n  (ty) -  t  y
b =
n t 2 - (  t) 2

 y - b t
a =
n

39

Linear Trend Equation Example

Week (t) Sales (y)

1 150
2 157
3 162
4 166
5 177
6 ??

Find forecast for week 6

40

20
 2/17/2020

Linear Trend Equation Example

Week (t) Sales (y) t2 ty

1 150
2 157
3 162
4 166
5 177

41

n  (ty) -  t  y
b =
n t 2 - (  t) 2

b = = =

a = 
y - b t
n
a = =

Ft = a + bt Ft = +
42

21
 2/17/2020

Find forecast for week 6

F6 =
=
=

43

Tutorial 3 Sept 2014

Canggih computer services and repairs personal computer at

its store. It primarily uses part time UiTM students as
technicians thus, they need a good forecasting of demand of
repairs so that they will know how many technicians to hire.
The company has accumulated the demand data in table
below.
Month Demand
Jan 37
Feb 40
Mac 41
Apr 37
May 45
Jun 50
July ??
Develop a linear trend line to forecast demand for July.

44

22
 2/17/2020

Linear Trend Calculation

Month Demand t t2 ty
(t) (y)
Jan 37
Feb 40
Mac 41
Apr 37
May 45
Jun 50

45

b=

a=

Ft = a + bt Ft =
Find forecast for July

F7 =

46

23
 2/17/2020

Tutorial 4

47

Tutorial 5

48

24
 2/17/2020

Tutorial 6

Assume 3 month Moving Average

Mac 2017

49

Semester Enrollment α=0.4 3MA

1
2
3
4
5
6

AD/Actual*100
1 - -
2
3
4
5
Sum
MAPE

50

25
 2/17/2020

Jun 2018

51

26