0% found this document useful (0 votes)
49 views10 pages

Sales Forecasting Using Moving Averages

The document provides examples of using different forecasting techniques, including weighted moving averages, exponential smoothing, and quantitative methods. Specifically, it calculates forecasts using 4-week and 2-week weighted moving averages, compares the errors between the two methods, and finds that the 2-week method produced a better forecast based on the mean absolute deviation (MAD) error measure. It also shows exponential smoothing forecasts using smoothing constants of 0.1 and 0.4, and calculates moving average forecasts for a data set using 3-period and 4-period moving averages.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
49 views10 pages

Sales Forecasting Using Moving Averages

The document provides examples of using different forecasting techniques, including weighted moving averages, exponential smoothing, and quantitative methods. Specifically, it calculates forecasts using 4-week and 2-week weighted moving averages, compares the errors between the two methods, and finds that the 2-week method produced a better forecast based on the mean absolute deviation (MAD) error measure. It also shows exponential smoothing forecasts using smoothing constants of 0.1 and 0.4, and calculates moving average forecasts for a data set using 3-period and 4-period moving averages.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

FORECASTING- WEIGHTED MOVING AVERAGES

QUESTION 1

- Calculate weighted moving averages


- Compare error measure using MAD
WEEK SALES
1 39
2 44
3 40
4 45
5 38
6 43
7 39
8

a) Forecast sales using 4-week weighted moving averages with weights 0.4,0.3,0.2 and 0.1
WEEK SALES 4WMA
1 39 -
2 44 -
3 40 -
4 45 -
5 38 42.7
6 43 41.1
7 39 41.6
8 40.6

Weights = 0.4+0.3+0.2+0.1 = 1 , 𝑊𝑀𝐴𝑛 = ∑𝑛𝑖−1 𝑊𝑖 𝐷𝑖

𝐹5 = 𝑊𝑀𝐴4 = (0.4)(45) + (0.3)(40) + (0.2)(44) + (0.1)(39)

= 42.7

𝐹6 = 𝑊𝑀𝐴4 = (0.4)(38) + (0.3)(45) + (0.2)(40) + (0.1)(44)

= 41.1

𝐹7 = 𝑊𝑀𝐴4 = (0.4)(43) + (0.3)(38) + (0.2)(45) + (0.1)(40)

= 41.6

𝐹8 = 𝑊𝑀𝐴4 = (0.4)(39) + (0.3)(43) + (0.2)(38) + (0.1)(45)

= 40.6
b) Forecast sales using 2-week weighted moving averages with weights 3 and 2
WEEK SALES 2WMA
1 39 -
2 44 -
3 40 42
4 45 41.6
5 38 43
6 43 40.8
7 39 41
8 40

Weights = 3+2 = 5 , 𝑊𝑀𝐴𝑛 = ∑𝑛𝑖−1 𝑊𝑖 𝐷𝑖

(3)(44) + (2)(39)
𝐹3 = 𝑊𝑀𝐴2 =
5

= 42

(3)(40) + (2)(44)
𝐹4 = 𝑊𝑀𝐴2 =
5

= 41.6

(3)(45) + (2)(40)
𝐹5 = 𝑊𝑀𝐴2 =
5

= 43

(3)(38) + (2)(45)
𝐹6 = 𝑊𝑀𝐴2 =
5

= 40.8

(3)(43) + (2)(38)
𝐹7 = 𝑊𝑀𝐴2 =
5

= 41

(3)(39) + (2)(43)
𝐹8 = 𝑊𝑀𝐴2 =
5

= 40
c) Compare forecast errors

WEEK SALES 4WMA |𝐷𝑡 − 𝐹𝑡 |


1 39 - -
2 44 - -
3 40 - -
4 45 - -
5 38 42.7 4.7
6 43 41.1 1.9
7 39 41.6 2.6
9.2

∑|𝐷𝑡 −𝐹𝑡 |
MAD =
𝑛
9.2
= 3
= 3.07
WEEK SALES 2WMA |𝐷𝑡 − 𝐹𝑡 |
1 39 - -
2 44 - -
3 40 42 2
4 45 41.6 3.4
5 38 43 5
6 43 40.8 2.2
7 39 41 2
14.6

∑|𝐷𝑡 −𝐹𝑡 |
MAD = 𝑛
14.6
=
5
= 2.92

Method MAD
4-week weighted average 3.07
2-week weighted average 2.92

Since MAD is the error measure, smaller MAD produce a better smoothing of the data. Therefore,
using MAD, the 2-week weighted average method produced a better forecast.
QUESTION 2

Use quantitative forecast methods for the data shown below.


Period 1 2 3 4 5 6 7
Observation 24 34 36 37 41 44 45

a) Compute for:

I. Naïve Method
Period 1 2 3 4 5 6 7 8
Observation 24 34 36 37 41 44 45
Forecast - 24 34 36 37 41 44 45

II. 3-period moving average


Period 1 2 3 4 5 6 7 8
Observation 24 34 36 37 41 44 45
Forecast - - - 31.33 35.67 38 40.67 43.33

∑𝑛𝑖=1 𝐷𝑖
𝑀𝐴𝑛 =
𝑛
36 + 34 + 24
𝐹4 = 𝑀𝐴3 =
3
= 31.33
37 + 36 + 34
𝐹5 = 𝑀𝐴3 =
3
= 35.67
41 + 37 + 36
𝐹6 = 𝑀𝐴3 =
3
= 38
44 + 41 + 37
𝐹7 = 𝑀𝐴3 =
3
= 40.67
45 + 44 + 41
𝐹8 = 𝑀𝐴3 =
3
= 43.33
III. 4-period moving average
Period 1 2 3 4 5 6 7 8
Observation 24 34 36 37 41 44 45
Forecast - - - - 32.75 37 39.5 41.75

∑𝑛𝑖=1 𝐷𝑖
𝑀𝐴𝑛 =
𝑛
37 + 36 + 34 + 24
𝐹5 = 𝑀𝐴4 =
4
= 32.75
41 + 37 + 36 + 34
𝐹6 = 𝑀𝐴4 =
4
= 37
44 + 41 + 37 + 36
𝐹7 = 𝑀𝐴4 =
4
= 39.5
45 + 44 + 41 + 37
𝐹8 = 𝑀𝐴3 =
4
= 41.75
FORECASTING- EXPONENTIAL SMOOTHING

QUESTION 3
PERIOD ACTUAL
1 42
2 40
3 43
4 40
5 41
6 39
7 46
8 44
9 45
10 38
11 40
12

a) Consider exponential smoothing forecasts using smoothing constants 0.1 and 0.4
PERIOD ACTUAL Forecast, 𝐹𝑡+1 (𝛼 = 0.1)
1 42 -
2 40 42
3 43 41.8
4 40 41.92
5 41 41.73
6 39 41.66
7 46 41.39
8 44 41.85
9 45 42.07
10 38 42.36
11 40 41.92
12 41.73

𝐹𝑡+1 = 𝛼𝐷𝑡 + (1 − 𝛼)𝐹𝑡

𝐹1+1 = 𝛼𝐷1 + (1 − 𝛼)𝐹1

𝐹2 = (0.1)(42) + (1 − 0.1)(42)

= 42

𝐹2+1 = 𝛼𝐷2 + (1 − 𝛼)𝐹2

𝐹3 = (0.1)(40) + (1 − 0.1)(42)

= 41.8

𝐹3+1 = 𝛼𝐷3 + (1 − 𝛼)𝐹3

𝐹4 = (0.1)(43) + (1 − 0.1)(41.8)
= 41.92

𝐹4+1 = 𝛼𝐷4 + (1 − 𝛼)𝐹4

𝐹5 = (0.1)(40) + (1 − 0.1)(41.92)

= 41.73

𝐹5+1 = 𝛼𝐷5 + (1 − 𝛼)𝐹5

𝐹6 = (0.1)(41) + (1 − 0.1)(41.73)

= 41.66

𝐹6+1 = 𝛼𝐷6 + (1 − 𝛼)𝐹6

𝐹7 = (0.1)(39) + (1 − 0.1)(41.66)

= 41.39

𝐹7+1 = 𝛼𝐷7 + (1 − 𝛼)𝐹7

𝐹8 = (0.1)(46) + (1 − 0.1)(41.39)

= 41.85

𝐹8+1 = 𝛼𝐷8 + (1 − 𝛼)𝐹8

𝐹9 = (0.1)(44) + (1 − 0.1)(41.85)

= 42.07

𝐹9+1 = 𝛼𝐷9 + (1 − 𝛼)𝐹9

𝐹10 = (0.1)(45) + (1 − 0.1)(42.07)

= 42.36

𝐹10+1 = 𝛼𝐷10 + (1 − 𝛼)𝐹10

𝐹11 = (0.1)(38) + (1 − 0.1)(42.36)

= 41.92

𝐹11+1 = 𝛼𝐷11 + (1 − 𝛼)𝐹11

𝐹12 = (0.1)(40) + (1 − 0.1)(41.92)

= 41.73
PERIOD ACTUAL Forecast, 𝐹𝑡+1 (𝛼 = 0.4)
1 42 -
2 40 42
3 43 41.2
4 40 41.92
5 41 41.15
6 39 41.09
7 46 40.25
8 44 43.15
9 45 43.49
10 38 44.09
11 40 41.65
12 40.99

𝐹𝑡+1 = 𝛼𝐷𝑡 + (1 − 𝛼)𝐹𝑡

𝐹1+1 = 𝛼𝐷1 + (1 − 𝛼)𝐹1

𝐹2 = (0.4)(42) + (1 − 0.4)(42)

= 42

𝐹2+1 = 𝛼𝐷2 + (1 − 𝛼)𝐹2

𝐹3 = (0.4)(40) + (1 − 0.4)(42)

= 41.2

𝐹3+1 = 𝛼𝐷3 + (1 − 𝛼)𝐹3

𝐹4 = (0.4)(43) + (1 − 0.4)(41.2)

= 41.92

𝐹4+1 = 𝛼𝐷4 + (1 − 𝛼)𝐹4

𝐹5 = (0.4)(40) + (1 − 0.4)(41.92)

= 41.15

𝐹5+1 = 𝛼𝐷5 + (1 − 𝛼)𝐹5

𝐹6 = (0.4)(41) + (1 − 0.4)(41.15)

= 41.09

𝐹6+1 = 𝛼𝐷6 + (1 − 𝛼)𝐹6

𝐹7 = (0.4)(39) + (1 − 0.4)(41.09)
= 40.25

𝐹7+1 = 𝛼𝐷7 + (1 − 𝛼)𝐹7

𝐹8 = (0.4)(46) + (1 − 0.4)(40.25)

= 43.15

𝐹8+1 = 𝛼𝐷8 + (1 − 𝛼)𝐹8

𝐹9 = (0.4)(44) + (1 − 0.4)(43.15)

= 43.49

𝐹9+1 = 𝛼𝐷9 + (1 − 𝛼)𝐹9

𝐹10 = (0.4)(45) + (1 − 0.4)(43.49)

= 44.09

𝐹10+1 = 𝛼𝐷10 + (1 − 𝛼)𝐹10

𝐹11 = (0.4)(38) + (1 − 0.4)(44.09)

= 41.65

𝐹11+1 = 𝛼𝐷11 + (1 − 𝛼)𝐹11

𝐹12 = (0.4)(40) + (1 − 0.4)(41.65)

= 40.99
Exponential Smoothing
PERIOD ACTUAL Forecast, 𝐹𝑡+1 (𝛼 = 0.1) Forecast, 𝐹𝑡+1 (𝛼 = 0.4)
1 42 - -
2 40 42 42
3 43 41.8 41.2
4 40 41.92 41.92
5 41 41.73 41.15
6 39 41.66 41.09
7 46 41.39 40.25
8 44 41.85 43.15
9 45 42.07 43.49
10 38 42.36 44.09
11 40 41.92 41.65
12 41.73 40.99

You might also like