RSH Qam11 Ism Ch05

CHAPTER 5
Forecasting
TEACHING SUGGESTIONS
Teaching Suggestion 5.1: Wide Use of Forecasting.
Forecasting is one of the most important tools a student can master because every firm needs to
conduct forecasts. Its useful to motivate students with the idea that obscure sounding techniques
such as exponential smoothing are actually widely used in business, and a good manager is
expected to understand forecasting. Regression is commonly accepted as a tool in economic and
legal cases.
Teaching Suggestion 5.2: Forecasting as an Art and a Science.
Forecasting is as much an art as a science. Students should understand that qualitative analysis
(judgmental modeling) plays an important role in predicting the future since not every factor can
be quantified. Sometimes the best forecast is done by seat-of-the-pants methods.
Teaching Suggestion 5.3: Use of Simple Models.
Many managers want to know what goes on behind the forecast. They may feel uncomfortable
with complex statistical models with too many variables. They also need to feel a part of the
process.
Teaching Suggestion 5.4: Management Input to the Exponential Smoothing Model.
One of the strengths of exponential smoothing is that it allows decision makers to input constants
that give weight to recent data. Most managers want to feel a part of the modeling process and
appreciate the opportunity to provide input.
Teaching Suggestion 5.5: Wide Use of Adaptive Models.
With todays dominant use of computers in forecasting, it is possible for a program to constantly
track the accuracy of a models forecast. Its important to understand that a program can
automatically select the best alpha and beta weights in exponential smoothing. Even if a firm has
10,000 products, the constants can be selected very quickly and easily without human
intervention.
Copyright 2012 Pearson Education, Inc. publishing as Prentice Hall
51
ALTERNATIVE EXAMPLES
Alternative Example 5.1:
Moving average
demand in previous n periods

n
Bicycle sales at Bowers Bikes are shown in the middle column of the following table. A 3-week
moving average appears on the right.
[ADDARROWSTOBELOWTABLE]
Actual
Three-Week
Week
Bicycle Sales
Moving Average
10
11
(8 + 10 + 9)/3 = 9
10
(10 + 9 + 11)/3 = 10
13
(9 + 11 + 10)/3 = 10
(11 + 10 + 13)/3 = 11 13
Alternative Example 5.2: Weighted moving average
weight for period n demand in period n

weights
Bowers Bikes decides to forecast bicycle sales by weighting the past 3 weeks as follows:
Weights Applied
Period
Last week
Two weeks ago
Three weeks ago
Sum of weights
A 3-week weighted moving average appears below.
52
Week
1
2
3
4
5
6
7
Actual Bicycle Sales

8
10
9
11
10
13
Three-Week Moving Average

[(3 9) + (2 10) + (1 8)]/6 = 9 1 6
[(3 11) + (2 9) + (1 10)]/6 = 10 1 6
[(3 10) + (2 11) + (1 9)]/6 = 10 1 6
[(3 13) + (2 10) + (1 11)]/6 = 11 2 3
Alternative Example 5.3: A firm uses simple exponential smoothing with = 0.1 to forecast
demand. The forecast for the week of January 1 was 500 units, whereas actual demand turned out
to be 450 units. The demand forecasted for the week of January 8 is calculated as follows.
Ft+1 = Ft + (Yt Ft)
= 500 + 0.1(450 500) = 495 units
Alternative Example 5.4: Exponential smoothing is used to forecast automobile battery sales.
Two values of are examined, = 0.8 and = 0.5. To evaluate the accuracy of each smoothing
constant, we can compute the absolute deviations and MADs. Assume that the forecast for
January was 22 batteries.
Month
January
February
March
April
May
June
Absolute
Actual
Forecast
Deviation
Battery
with
With
Sales
= 0.8
= 0.8
20
22
2
21
20.40
0.6
15
20.880
5.88
14
16.176
2.176
13
14.435
1.435
16
13.287
2.713
Sum of absolute deviations: 14.804
MAD:
2.467
Forecast
with
= 0.5
22
21
21
18
16
14.5
Absolute
Deviation
with
= 0.5
2
0
6
4
3
1.5
16.5
2.75
On the basis of this analysis, a smoothing constant of = 0.8 is preferred to = 0.5 because it
has a smaller MAD.
53
Alternative Example 5.5: Use the sales data given below to determine: (a) the least squares
trend line, (b) the predicted value for 2010 sales.
Time
Period
1
2
3
4
5
6
7
X = 28
Year
2003
2004
2005
2006
2007
2008
2009
X2
1
4
9
16
25
36
49
2
X = 140
XY
100
220
366
520
695
912
1,148
XY= 3,961
28
Y 917 131
4
Y
n
7
n
7
XY n X Y 3,961 7 4 131 293 10.464
b
2
28
140 7 42
X 2 nX
X
Sales
(Units)
100
110
122
130
139
152
164
Y = 917
a Y b X 131 10.46 4 89.14
The trend equation is

^
Y b0 b1 X 89.14 10.464 X
To project demand in 2010, we denote the year 2010 as x = 8,
Sales in 2000 = 89.14 + 10.464(8) = 172.85
Alternative Example 5.6: The rated power capacity (in hours/ week) over the past 6 years is
shown in the table below.
Year
Capacity
(Y)
X2
XY
115
120
118
4
5
6
X = 21
124
123
130
Y = 730
16
25
36
X2 = 91
496
615
780
XY = 2600
54
X 21/ 6 3.5
Y 730 / 6 121.667
XY n XY 2600 6(3.5)(121.667) 2.57

91 6(3.5)
X nX
Y b X 730 2.57(3.5) 112.67
b1
b0
Y 112.67 2.57X
Forecast for year 7 = 112.67 + (2.57)(7)
=130.7
Alternative Example 5.7: The forecast demand and actual demand for 10-foot fishing boats are
shown below. We compute the tracking signal and MAD.
MAD
Forecast errors 70 11.7

n
Tracking Signal
RSFE 24
2.1 MADs
MAD 11.7
Table for Alternate Example 5.7

Year
1
2
3
4
5
6
Forecast Actual
Demand Demand
78
71
75
80
83
101
84
84
88
60
85
73
Error
7
5
18
0
28
12
RSFE
7
2
16
16
12
24
Forecast Cumulative
Error
Error
7
7
5
12
18
30
0
30
28
58
12
70
MAD
7.0
6.0
10.0
7.5
11.6
11.7
Tracking
Signal
1.0
0.3
+1.6
+2.1
1.0
2.1
55
SOLUTIONS TO DISCUSSION QUESTIONS AND PROBLEMS

5-1. The steps that are used to develop any forecasting system are:
1. Determine the use of the forecast.
2. Select the items or quantities that are to be forecasted.
3. Determine the time horizon of the forecast.
4. Select the forecasting model.
5. Gather the necessary data.
6. Validate the forecasting model.
7. Make the forecast.
8. Implement the results.
5-2. A time-series forecasting model uses historical data to predict future trends.
5-3. The only difference between causal models and time-series models is that causal models take
into account any factors that may influence the quantity being forecasted. Causal models use
historical data as well. Time-series models use only historical data.
5-4. Qualitative models incorporate subjective factors into the forecasting model. Judgmental
models are useful when subjective factors are important. When quantitative data are difficult to
obtain, qualitative models are appropriate.
5-5. The disadvantages of the moving average forecasting model are that the averages always
stay within past levels, and the moving averages do not consider seasonal variations.
5-6. When the smoothing value, , is high, more weight is given to recent data. When is low,
more weight is given to past data.
5-7. The Delphi technique involves analyzing the predictions that a group of experts have made,
then allowing the experts to review the data again. This process may be repeated several times.
After the final analysis, the forecast is developed. The group of experts may be geographically
dispersed.
5-8. MAD is a technique for determining the accuracy of a forecasting model by taking the
average of the absolute deviations. MAD is important because it can be used to help increase
forecasting accuracy.
5-9. The number of seasons depends on the number of time periods that occur before a pattern
repeats itself. For example, monthly data would have 12 seasons because there are 12 months in a
year. Quarterly data would have 4 seasons because there are 4 quarters in a year. Daily data
would have 7 seasons because there are 7 days in a week. For daily data, it is common for many
retail stores to have higher sales on Saturdays than on other days of the week, and a seasonal
index would reflect that.
56
5-10. If a seasonal index equals 1, that season is just an average season. If the index is less than 1,
that season tends to be lower than average. If the index is greater than 1, that season tends to be
higher than average.
5-11. If the smoothing constant equals 0, then
Ft+1 = Ft + 0(Yt Ft) = Ft
This means that the forecast never changes.
If the smoothing constant equals 1, then
Ft+1 = Ft + 1(Yt Ft) = Yt
This means that the forecast is always equal to the actual value in the prior period.
5-12. A centered moving average (CMA) should be used if trend is present in data. If an overall
average is used rather than a CMA, variations due to trend will be interpreted as variations due to
seasonal factors. Thus, the seasonal indices will not be accurate.
5-13.
Actual
Month
Shed Sales
Four-Month Moving Average
Jan.
10
Feb.
12
Mar.
13
Apr.
16
May
19
June
23
July
26
Aug.
30
(16 + 19 + 23 + 26)/4 = 84/4 = 21
Sept.
28
(19 + 23 + 26 + 30)/4 = 98/4 = 24.5
Oct.
18
Nov.
16
Dec.
14
(10 + 12 + 13 + 16)/4 = 51/4 = 12.75

(12 + 13 + 16 + 19)/4 = 60/4 = 15
(13 + 16 + 19 + 23)/4 = 70/4 = 17.75
(23 + 26 + 30 + 28)/4 = 107/4 =

26.75
(26 + 30 + 28 + 18)/4 = 102/4 = 25.5
(30 + 28 + 18 + 16)/4 = 92/4 = 23
The MAD = 7.78

See solution to 5-13 for calculations.
57
5-14.
Three-
Month
Four-
Three-
Month
Four-
Month
Actual
Month
Absolute
Month
Absolute
Shed Sales
Forecast
Deviation
Forecast
Deviation
Jan.
10
Feb.
12
Mar.
13
Apr.
16
11.67
4.33
May
19
13.67
5.33
12.75
6.25
June
23
16
15
July
26
19.33
6.67
17.75
8.25
Aug.
30
22.67
7.33
21
Sept.
28
26.33
1.67
24.5
3.5
Oct.
18
28
26.75
8.75
Nov.
16
25.33
9.33
25.5
9.5
Dec.
14
20.67
6.67
23
10
58.33
Three-month MAD
Four-month MAD
62.25
58.33
6.48
9
62.25
7.78
8
The 3-month moving average appears to be more accurate. However, when weighted moving
averages were used, the MAD was 5.444.
5-15.
58
Year
Demand
3-Year Moving
Ave.
(4 + 6 + 4)/3
= 4 23
10
(6 + 4 + 5)/3
=5
(4 + 5 + 10)/3
= 6 13
(5 + 10 + 8)/3
= 7 23
(10 + 8 + 7)/3
= 8 13
12
(8 + 7 + 9)/3
=8
10
14
(7 + 9 + 12)/3
= 9 13
11
15 (9 + 12 + 14)/3
=11 2 3
3-Year Wt. Moving

Ave.
3-Year Abs.
Deviation
3-Year Wt. Abs.

Deviation
[(2 4) + 6 + 4]/4
= 4 12
0.34
0.55
[(2 5) + 4 + 6]/4
=5
[(2 10) + 5 +4]/4

= 7 14
1.67
0.75
[(2 8) + 10 +5]/4
= 7 34
0.67
0.75
[(2 7) + 8 +10]/4
=8
0.67
3.75
4.67
3.34
2.75
20.36
18.55
[(2 9) + 7 + 8]/4
= 8 14
[(2 12) + 9 +7]/4
= 10
[(2 14) + 12+9]/4
= 12 1 4
Total absolute deviations:
MAD for 3-year average = 20.36/8 = 2.55

MAD for weighted 3-year average = 18.55/8 = 2.32
The weighted moving average appears to be slightly more accurate in its annual forecasts.
59
5-16. Using Excel or QM for Windows, the trend line is

Y = 2.22 +1.05X
Where X = time period (1, 2, . . ., 11) Y = demand
5-17. Using the forecasts in the previous problem we obtain the absolute deviations given in the
table below.
Year
Demand
3-Yr MA
3-Yr Wt.
MA
Trend line
|deviation|
|deviation|
|deviation|
0.73
1.67
1.38
0.34
0.55
1.44
10
5.00
5.00
2.51
1.67
0.75
0.55
0.67
0.75
2.60
0.67
1.00
1.65
12
4.00
3.75
0.29
10
14
4.67
4.00
1.24
11
15
3.34
2.75
1.18
20.36
18.55
15.24
Total absolute deviations =
MAD (3-year moving average) = 2.55

MAD (3-year weighted moving average) = 2.32
MAD (trend line) = 1.39
The trend line is best because the MAD for that method is lowest.
5-18. = 0.3. New forecast for year 2 is last periods forecast + (last periods actual demand
last periods forecast):
new forecast for year 2 = 5,000 + (0.3)(4,000 5,000)
= 5,000 + (0.3)( 1,000)
= 5,000 300
= 4,700
510
The calculations are:

Year
Demand
New Forecast
6,000
4,700 = 5,000 + (0.3)(4,000 5,000)
4,000
5,090 = 4,700 + (0.3)(6,000 4,700)
5,000
4,763 = 5,090 + (0.3)(4,000 5,090)
10,000
4,834 = 4,763 + (0.3)(5,000 4,763)
8,000
6,384 = 4,834 + (0.3)(10,000 4,834)
7,000
6,869 = 6,384 + (0.3)(8,000 6,384)
9,000
6,908 = 6,869 + (0.3)(7,000 6,869)
12,000
7,536 = 6,908 + (0.3)(9,000 6,908)
10
14,000
8,875 = 7,536 + (0.3)(12,000 7,536)
11
15,000
10,412 = 8,875 + (0.3)(14,000 8,875)
The mean absolute deviation (MAD) can be used to determine which forecasting method is more
accurate.
Weighted
Year
Demand
Moving
Absolute
Average
Deviation
Absolute
Exp. Sm.
Deviation
4,000
5,000
1,000
6,000
4,700
1,300
4,000
5,090
1,090
5,000
4,500
500
4,763
237
10,000
5,000
5,000
4,834
5,166
8,000
7,250
750
6,384
1,616
7,000
7,750
750
6,869
131
9,000
8,000
1,000
6,908
2,092
12,000
8,250
3,750
7,536
4,464
10
14,000
10,000
4,000
8,875
5,125
11
15,000
12,250
2,750
10,412
4,588
Total:
18,500
Mean:
2,312.5
26,808
2,437
Thus, the 3-year weighted moving average model appears to be more accurate.
511
5-19. = 0.30
Year
Forecast
5-20.
Year
1
2
3
4
5
6
Sales
450
495
518
563
584
?
410.0
422.0
443.9
466.1
495.2
521.8
Forecast Using = 0.6

410
410 + (0.6)(450 410) = 434
434 + (0.6)(495 434) = 470.6
470.6 + (0.6)(518 470.6) = 499.0
499 + (0.6)(563 499) = 537.4
537.4 + (0.6)(584 537.4) = 565.4
Forecast Using = 0.9

410 + (0.9)(450 410) = 446
446 + (0.9)(495 446) = 490.1
490.1 + (0.9)(518 490.1) = 515.21
515.21 + (0.9)(563 515.21) = 558.2
558.2 + (0.9)(584 558.2) = 581.4
5-21.
Year
1
2
3
4
5
6
Actual
Sales
= 0.3
Forecast
Absolute
Deviatio
n
450
410.0
40.0
495
422.0
73.0
518
443.9
74.1
563
466.1
96.9
584
495.2
88.8
?
521.8
Total absolute deviation 372.8
= 0.6
Forecast
410.0
434.0
470.6
499.0
537.4
565.4
Absolute
Deviatio
n
40.0
61.0
47.4
64.0
46.6
259.0
= 0.9
Forecast
410.0
446.0
490.1
515.2
558.2
581.4
Absolute
Deviatio
n
40.0
49.0
27.9
47.8
25.8
190.5
MAD=0.3 = 372.8/5 = 74.56

MAD=0.6 = 259/5 = 51.8
MAD=0.9 = 190.5/5 = 38.1
Because it has the lowest MAD, the smoothing constant = 0.9 gives the most accurate forecast.
5-22.
Year
1
2
3
4
5
6
Sales
450
495
518
563
584
?
Three-Year Moving Average
(450 + 495 + 518)/3 = 487.67

(495 + 518 + 563)/3 = 525.3
(518 + 563 + 584)/3 = 555
512
5-23.
Time
Period
X
1
2
3
4
5
Year
1
2
3
4
5
Sales
Y
450
495
518
563
584
2,610
X2
1
4
9
16
25
55
XY
450
990
1554
2252
2920
8166
b1 = 33.6
b0 = 421.2
Y = 421.2 + 33.6X
Projected sales in year 6,
Y = 421.2 + (33.6)(6)
= 622.8
5-24.
Year
1
2
3
4
5
6
Actual Sales
Three-Year Moving
Average Forecast
450
495
518
563
487.7
584
525.3
?
555.0
Total absolute deviation
MAD=0.3 = 74.56
Absolute
Deviation
75.3
58.7
134.0
Time-Series
Forecast
454.8
488.4
522.0
555.6
589.2
622.8
Absolute
Deviation
4.8
6.6
4.0
7.4
5.2
28.0
(see Problem 5-21)
MADmoving average = 134/2 = 67

MADregression = 28/5 = 5.6
Regression (trend line) is obviously the preferred method because of its low MAD.
513
5-25. To answer the discussion questions, two forecasting models are required: a three-period
moving average and a three-period weighted moving average. Once the actual forecasts have been
made, their accuracy can be compared using the mean absolute differences (MAD).
a., b. Because a three-period average forecasting method is used, forecasts start for period 4.
Period
4
5
6
7
8
9
10
11
12
13
14
Month
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Jan.
Feb.
Demand
10
15
17
11
14
17
12
14
16
11
Average
13.67
13.33
13.67
14
14.33
14
14
14.33
14.33
14
13.67
Weighted Average
14.5
12.67
13.5
15.17
13.67
13.50
15
14
13.83
14.67
13.17
c. MAD for moving average is 2.2. MAD for weighted average is 2.72. Moving average
forecast for February is 13.67. Weighted moving average forecast for February is 13.17.
Thus, based on this analysis, the moving average appears to be more accurate. The forecast
for February is about 14.
d. There are many other factors to consider, including seasonality and any underlying causal
variables such as advertising budget.
5-26. a. = 0.20
Week
1
2
3
4
5
6
7
8
9
10
11
12
Actual
Miles
17
21
19
23
18
16
20
18
22
20
15
22
Forecast
(Ft)
17.00
17.00
17.80
18.04
19.03
18.83
18.26
18.61
18.49
19.19
19.35
18.48
Error
+4.00
+1.20
+4.96
1.03
2.83
+1.74
0.61
+3.51
+0.81
4.35
+3.52
RSFE
+4.00
+5.20
+10.16
+9.13
+6.30
+8.04
+7.43
+10.94
+11.75
+7.40
+10.92
Sum of
Absolute
Forecast
Errors
4.00
5.20
10.16
11.19
14.02
15.76
16.37
19.88
20.69
25.04
28.56
MAD
4.00
2.60
3.39
2.80
2.80
2.63
2.34
2.49
2.30
2.50
2.60
Track Signal
1
2
3
3.3
2.25
3.05
3.17
4.21
5.11
2.96
4.20
b. The total MAD is 2.60.

c. RSFE is consistently positive. Tracking signal exceeds 2 MADs at week 10. This could
indicate a problem.
514
5-27. a., b. See the accompanying table for a comparison of the calculations for the exponentially
smoothed forecasts using constants of 0.1 and 0.6.
c. Students should note how stable the smoothed values for the 0.1 smoothing constant are.
When compared to actual week 25 calls of 85, the 0.6 smoothing constant appears to do a
better job. On the basis of the forecast error, the 0.6 constant is better also. However, other
smoothing constants need to be examined.
Actual
Week,
Value,
t
Yt
1
50
2
35
3
25
4
40
5
45
6
35
7
20
8
30
9
35
10
20
11
15
12
40
13
55
14
35
15
25
16
55
17
55
18
40
19
35
20
60
21
75
22
50
23
40
24
65
25
Smoothed
Value,
Ft( = 0.1)
50
50.00
48.50
46.15
45.54
45.48
44.43
41.99
40.79
40.21
38.19
35.87
36.28
38.16
37.84
36.56
38.40
40.06
40.05
39.55
41.59
44.93
45.44
44.90
46.91
Smoothed
Forecast
Value,
Forecast
Error
Error
Ft( = 0.6)
-15.00
50.00
-15.00
-23.50
41.00
-16.00
-6.15
31.40
8.60
-0.54
36.56
8.44
-10.48
41.62
-6.62
-24.43
37.65
-17.65
-11.99
27.06
2.94
-5.79
28.82
6.18
-20.21
32.53
-12.53
-23.19
25.01
-10.01
4.13
19.00
21.00
18.72
31.60
23.40
-3.16
45.64
-10.64
-12.84
39.26
-14.26
18.44
30.70
24.30
16.60
45.28
9.72
-0.06
51.11
-11.11
-5.05
44.45
-9.45
20.45
38.78
21.22
33.41
51.51
23.49
5.07
65.60
-15.60
-5.44
56.24
-16.24
20.10
46.50
18.50
57.60
515
5-28. Using data from Problem 5-27, with = 0.9

Week,
t
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Actual
Value,
Yt
50
35
25
40
45
35
20
30
35
20
15
40
55
35
25
55
55
40
35
60
75
50
40
65
Smoothed
Value,
Ft( = 0.9)
50
50.00
36.50
26.15
38.62
44.36
35.94
21.59
29.16
34.42
21.44
15.64
37.56
53.26
36.83
26.18
52.12
54.71
41.47
35.65
57.56
73.26
52.33
41.23
62.62
Forecast
Error
-15.00
-11.50
13.85
6.39
-9.36
-15.94
8.41
5.84
-14.42
-6.44
24.36
17.44
-18.26
-11.83
28.82
2.88
-14.71
-6.47
24.35
17.44
-23.26
-12.33
23.77
Note that in this problem, the initial forecast (for the first period) was not used in computing the
MAD = 14.48. Either approach is considered valid.
516
5-29. Exponential smoothing with = 0.1

Month
Feb.
March
April
May
June
July
Aug.
Income
70.0
68.5
64.8
71.7
71.3
72.8
Forecast
Error
65.0
3.0
65.0 + 0.1(70 65) = 65.5
65.5 + 0.1(68.5 65.5) = 65.8
1.0
6.0
65.8 + 0.1(64.8 65.8) = 65.7
5.0
65.7 + 0.1(71.7 65.7) = 66.3
6.0
66.3 + 0.1(71.3 66.3) = 66.8
66.8 + 0.1(72.8 66.8) = 67.4
MAD = 4.20
MAD. Either approach is considered valid.
5-30. Exponential smoothing with = 0.3
Month
Feb.
March
April
May
June
July
Aug.
Income
70.0
68.5
64.8
71.7
71.3
72.8
Forecast
Error
65.0
66.5
2.0
67.1
2.3
66.4
5.3
68.0
3.3
69.0
3.8
70.1
MAD = 3.34
Based on MAD, = 0.3 produces a better forecast than = 0.1 (of Problem 5-29).
MAD. Either approach is considered valid.
5-31. Using QM for Windows, we select Forecasting - Time Series and multiplicative
decomposition. Then specify Centered Moving Average and we have the following results:
a. Quarter 1 index = 0.8825; Quarter 2 index = 0.9816; Quarter 3 index = 0.9712; Quarter 4
index = 1.1569
b. The trendline is Y = 237.7478 + 3.6658X
c. Quarter 1: Y = 237.7478 + 3.6658(17) = 300.0662
Quarter 2: Y = 237.7478 + 3.6658(18) = 303.7320
Quarter 3: Y = 237.7478 + 3.6658(19) = 307.3978
Quarter 4: Y = 237.7478 + 3.6658(20) = 311.0636
517
d. Quarter 1: 300.0662(0.8825) = 264.7938

Quarter 2: 303.7320(0.9816) = 298.1579
Quarter 3: 307.3978(0.9712) = 298.5336
Quarter 4: 311.0636(1.1569) = 359.8719
5-32. Letting
t = time period (1, 2, 3, . . . , 16)
Q1 = 1 if quarter 1, 0 otherwise
Note: if Q1 = Q2 = Q3 = 0, then it is quarter 4.
Using computer software we get
Y = 281.6 + 3.7t 75.7Q1 48.9Q2 52.1Q3
The forecasts for the next 4 quarters are:
Y = 281.6 + 3.7(17) 75.7(1) 48.9(0) 52.1(0) = 268.7
Y = 281.6 + 3.7(18) 75.7(0) 48.9(1) 52.1(0) = 299.2
Y = 281.6 + 3.7(19) 75.7(0) 48.9(0) 52.1(1) = 299.7
Y = 281.6 + 3.7(20) 75.7(0) 48.9(0) 52.1(0) = 355.4
5-33 a. Using computer software we get Y = 197.5 0.34X where X = time period.
The slope is -0.34 which indicates a small negative trend. Note that the results are not statistically
significant and r2 = 0.001
b) Using QM for Windows for the multiplicative decomposition method with 4 seasons asnd
using a centered moving average, the seasonal indices are 1.47, 0.96, 0.70, and 0.87 for quarters
1-4 respectively. The trend equation found with the deseasonalized data is Y = 176.63+ 2.20X.
The slope of 2.20 indicates a positive trend of 2.20 per time period. The results are statistically
significant.
c) The negative slope of the trend line in part (a) was found when the seasonality was ignored.
The first quarter has a high seasonal index, so the first observation was very large relative to the
last observation. Thus, by looking at the raw data, which was used for the trend line in part (a), it
appeared that there was a negative trend but in reality this was due to the seasonal variations and
not due to trend. The decomposition method is better to use when there is a seasonal pattern
present.
518
5-34. For a smoothing constant of 0.2, the forecast for year 11 is 6.489.
Year
1
2
3
4
5
6
7
8
9
10
11
Rate
7.2
7
6.2
5.5
5.3
5.5
6.7
7.4
6.8
6.1
Forecast
|Error|
7.2
0
7.2
0.2
7.16
0.96
6.968
1.468
6.674
1.374
6.400
0.900
6.220
0.480
6.316
1.084
6.533
0.267
6.586
0.486
6.489
MAD = 0.722
FOR A SMOOTHING CONSTANT OF 0.4, THE FORECAST FOR YEAR 11 IS 6.458.

Year
1
2
3
4
5
6
7
8
9
10
11
Rate
7.2
7
6.2
5.5
5.3
5.5
6.7
7.4
6.8
6.1
Forecast
7.2
7.2
7.12
6.752
6.251
5.871
5.722
6.113
6.628
6.697
6.458
|Error|
0
0.2
0.92
1.252
0.951
0.371
0.978
1.287
0.172
0.597
MAD = 0.673

Year
1
2
3
4
5
6
7
8
9
10
11
Rate
7.2
7
6.2
5.5
5.3
5.5
6.7
7.4
6.8
6.1
Forecast
7.2
7.2
7.08
6.552
5.921
5.548
5.519
6.228
6.931
6.852
6.401
|Error|
0
0.2
0.88
1.052
0.621
0.048
1.181
1.172
0.131
0.752
MAD = 0.604

Year
Rate
Forecast
|Error|
519
1
2
3
4
5
6
7
8
9
10
11
7.2
7
6.2
5.5
5.3
5.5
6.7
7.4
6.8
6.1
7.2
7.2
7.04
6.368
5.674
5.375
5.475
6.455
7.211
6.882
6.256
0
0.2
0.84
0.868
0.374
0.125
1.225
0.945
0.411
0.782
MAD = 0.577
THE LOWEST MAD IS 0.577 FOR A SMOOTHING CONSTANT OF 0.8.

5-35. To compute a seasonalized or adjusted sales forecast, we just multiply each seasonal index
by the appropriate trend forecast.
= seasonal index trend forecast
Hence for:
Quarter I: I = (1.30)($100,000) = $130,000
Quarter II: II = (0.90)($120,000) = $108,000
Quarter III: III = (0.70)($140,000) = $98,000
Quarter IV: IV = (1.10)($160,000) = $176,000
5-36.
(Average demand for season)
Overall average demand =
Season index
year 1 demand year 2 demand

2
sum of all values

8
average for season
overall average demand

new annual demand
Year 3 demand
4
1, 200
season index
4
520
Solution Table for Problem 5-36
Season
Fall
Winter
Spring
Summer
Year 1
Demand
200
350
150
300
Year 2
Demand
250
300
165
285
(Average Year 1Year 2 Demand)

225.0
325.0
157.5
292.5
Average
Season
Demand
250
250
250
250
Season
Index
0.90
1.30
0.63
1.17
Year 3
Demand
270
390
189
351
5-37. Using Excel with X = 1, 2, 3, , 20 for years 1991-2010 respectively, the trend equation is
Y = 2898.6 + 499.7X.
For 2009, X = 21; Y = 2898.6 + 499.7 (21) = 13392
For 2010, X = 22; Y = 2898.6 + 499.7 (22) = 13892
For 2011, X = 23; Y = 2898.6 + 499.7 (23) = 14391
The MSE from the Excel output is 2740276.
5-38. Using QM for Windows, the forecast is 10229 and the MSE = 2676625 (ignoring the first
error). This MSE is lower than the MSE found using a trend line, so this technique worked better
than the trend line.
5-39. a. With a smoothing constant of 0.4, the forecast for 2011 is 10609 with MSE = 3703109.
(ignoring the first error)
b. Using QM for Windows, the best smoothing constant is 0.92. This gives the lowest MSE
of 2514990.
5-40. Using Excel, the trend equation is Y = 1.2672 + 0.01938X.
For January of 2010, X = 13; Y = 1.2672 + 0.01938(13) = 1.519.
For February of 2010, X = 14; Y = 1.2672 + 0.01938(14) = 1.538.
5-41. The forecast for January 2010 would be 1.452.
The MSE with the trend equation is 0.00049. The MSE with this exponential smoothing model is
0.00268.
SOLUTIONS TO INTERNET HOMEWORK PROBLEMS
5-42. With = 0.4, forecast for 2011 = 10338 and MAD = 836 (including first error). With =
0.6, forecast for 2011 = 10697 and MAD = 612.
5-43. Using Excel, the trend line is: GDP = 6,142.7 + 441.4(time). For 2011 (time = 12) the
forecast is GDP = 6,142.7 + 441.4 (12) = 11,4389.5.
521
5-44. The trend line found using Excel is: Patients = 29.73 + 3.28(time). Note these coefficients are
rounded. For the next 3 years (time = 11, 12, and 13) the forecasts for the number of patients are:
Patients = 29.73 + 3.28(11) = 65.8
Patients = 29.73 + 3.28(12) = 69.1
Patients = 29.73 + 3.28(13) = 72.4
The coefficient of determination is 0.85, so the model is a fair model.
5-45. The trend line found using Excel is: Crime Rate = 51.98 + 6.09(time). Note these coefficients
are rounded. For the next 3 years (time = 11, 12, and 13) the forecasts for the crime rates are:
Crime Rate = 51.98 + 6.09(11) = 118.97
Crime Rate = 51.98 + 6.09(12) = 125.06
Crime Rate = 51.98 + 6.09(13) = 131.15
The coefficient of determination is 0.96, so this is a very good model.
5-46. The regression equation (from Excel) is: Patients = 1.23 + 0.54(crime rate). Note these
coefficients are rounded. If the crime rate is 131.2, the forecast number of patients is:
Patients = 1.23 + 0.54(131.2) = 72.1
If the crime rate is 90.6, the forecast number of patients is:
Patients = 1.23 + 0.54(90.6) = 50.2
The coefficient of determination is 0.90, so this is a good model.
5-47. With = 0.6, forecast for 2003 = 86.2 and MAD = 3.42. With = 0.2, forecast for 2003 =
63.87 and MAD = 7.23. The model with = 0.6 is better since it has a lower MAD.
5-48. With = 0.6, forecast for 2003 = 4.86 and MAD = 0.23. With = 0.2, forecast for 2003 =
4.52 and MAD = 0.48. The model with = 0.6 is better since it has a lower MAD.
5-49. The trend line (coefficients from Excel are rounded) for deposits is:
Deposits = 18.968 + 1.638(time)
For 2003, 2004, and 2005, time = 45, 46, and 47 respectively. The forecasts are:
Deposits = 18.968 + 1.638(45) = 54.7
Deposits = 18.968 + 1.638(46) = 56.4
Deposits = 18.968 + 1.638(47) = 58.0
The trend line (coefficients from Excel are rounded) for GSP is:
GSP = 0.090 + 0.112(time). The forecasts are:
GSP = 0.090 + 0.112(45) = 5.1
GSP = 0.090 + 0.112(46) = 5.2
GSP = 0.090 + 0.112(47) = 5.4
5-50. The regression equation from Excel is
522
Deposits = 17.64 + 13.59(GSP)

In the scatterplot of this data that follows, the pattern appears to change around 1985. There are
definitely different relationships before 1985 and after 1985, so perhaps the model should be
developed with 1985 as the first year of data.
CASE STUDIES
FORECASTING ATTENDANCE AT SWU FOOTBALL GAMES
1. Because we are interested in annual attendance and there are six years of data, we find the
average attendance in each year shown in the table below. A graph of this indicates a linear
trend in the data. Using Trend Analysis in the forecasting module of QM for Windows we find
the equation:
Y = 31,660 + 2,305.714X
Where Y is attendance and X is the time period (X = 1 for 2005, 2 for 2006, etc.).
For this model, r2 = 0.98 which indicates this model is very accurate.
Attendance in 2011 is projected to be
523
Y = 31,660 + 2,305.714(7) = 47,800

Attendance in 2012 is projected to be
Y = 31,660 + 2,305.714(8) = 50,105
At this rate, the stadium, with a capacity of 54,000, will be maxed out (filled to capacity) in
2014.
Year
2005
2006
2007
2008
2009
2010
Attendance
34840
35380
38520
40500
43320
45820
2. Based upon the projected attendance and tickets prices of $20 in 2011 and $21 (a 5%
increase) in 2012, the projected revenues are:
47,800(20) = $956,000 in 2010 and
50,105(21) = $1,052,205 in 2012.
3. The school might consider another expansion of the stadium, or raise the ticket prices more
than 5% per year. Another possibility is to raise the prices of the best seats while leaving the
end zone prices more reasonable.
FORECASTING MONTHLY SALES

1.
The scatter plot of the data shows a definite seasonal pattern with higher sales in the winter
months and lower sales in the summer and fall months. There is a slight upward trend as
evidenced by the fact that for each month, the sales increased from the first year to the second,
and again form the second year to the third.
524
2. A trend line based on the raw data is found to be:

Y = 330.889 1.162X
The slope of the trend line is negative which would indicate that sales are declining over time.
However, as previously noted, sales are increasing. The high seasonal index in January and
February causes the trend line on the unadjusted data to appear to have a negative slope.
3. There is a definite seasonal pattern and a definite trend in the data. Using the decomposition
method in QM for Windows, the trend equation (based on the deseasonalized data) is
Y = 294.069 + 0.859X
The table below gives the seasonal indices, the unadjusted forecasts found using the trend line,
and the final (adjusted) forecasts for the next year.
Month
January
February
March
April
May
June
July
August
September
October
November
December
Unadjusted forecast
325.852
326.711
327.57
328.429
329.288
330.147
331.006
331.865
332.724
333.583
334.442
335.301
Seasonal index
1.447
1.393
1.379
1.074
1.039
0.797
0.813
0.720
0.667
0.747
0.891
1.033
Adjusted forecast
471.5
455.1
451.7
352.7
342.1
263.1
269.1
238.9
221.9
249.2
298.0
346.4
SOLUTION TO INTERNET CASES

SOLUTION TO AKRON ZOOLOGICAL PARK CASE
1. The instructor can use this question to have the student calculate a simple linear regression,
using real-world data. The attendance would be the dependent variable and time would be the
independent variable. From the attendance, the expected revenues could be determined. Also, the
instructor can broaden this question to include several other forecast techniques. For example,
exponential smoothing, last-period demand, or n-period moving averages can be assigned. It can
be explained that mean absolute deviation (MAD) is one of but a few methods by which analysts
can select the more appropriate forecast technique and outcome.
525
First, we perform a linear regression with time as the independent variable. The model that
results is
admissions = 44,352 + 9,197 year
(where year is coded as 1 = 1989, 2 = 1990, etc.)
r = 0.88
MAD = 9,662
MSE = 201,655,824
So the forecasts for 1999 and 2000 are 145,519 and 154,716, respectively. Using a weighted
average of $2.875 to represent gate receipts per person, revenues for 1999 and 2000 are
$418,367 and $444,808, respectively.
Students could also consider the impact of the increasing fees to see if the increase had an
impact on attendance.
2. The student should respond that the other factors are the variability of the weather, the special
events, the competition, and the role of advertising.
526

RSH Qam11 Ism Ch05

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

RSH Qam11 Ism Ch05

Uploaded by

Copyright:

Available Formats

CHAPTER 5

Copyright 2012 Pearson Education, Inc. publishing as Prentice Hall

demand in previous n periods

Alternative Example 5.2: Weighted moving average

weight for period n demand in period n

Two weeks ago

Three weeks ago

A 3-week weighted moving average appears below.

Copyright 2012 Pearson Education, Inc. publishing as Prentice Hall

Actual Bicycle Sales

Three-Week Moving Average

Copyright 2012 Pearson Education, Inc. publishing as Prentice Hall

a Y b X 131 10.46 4 89.14

The trend equation is

Copyright 2012 Pearson Education, Inc. publishing as Prentice Hall

XY n XY 2600 6(3.5)(121.667) 2.57

Forecast errors 70 11.7

Table for Alternate Example 5.7

Copyright 2012 Pearson Education, Inc. publishing as Prentice Hall

SOLUTIONS TO DISCUSSION QUESTIONS AND PROBLEMS

Copyright 2012 Pearson Education, Inc. publishing as Prentice Hall

Four-Month Moving Average

(16 + 19 + 23 + 26)/4 = 84/4 = 21

(19 + 23 + 26 + 30)/4 = 98/4 = 24.5

(10 + 12 + 13 + 16)/4 = 51/4 = 12.75

(23 + 26 + 30 + 28)/4 = 107/4 =

The MAD = 7.78

Copyright 2012 Pearson Education, Inc. publishing as Prentice Hall

Copyright 2012 Pearson Education, Inc. publishing as Prentice Hall

3-Year Wt. Moving

3-Year Wt. Abs.

[(2 10) + 5 +4]/4

Total absolute deviations:

MAD for 3-year average = 20.36/8 = 2.55

Copyright 2012 Pearson Education, Inc. publishing as Prentice Hall

5-16. Using Excel or QM for Windows, the trend line is

Total absolute deviations =

MAD (3-year moving average) = 2.55

Copyright 2012 Pearson Education, Inc. publishing as Prentice Hall

The calculations are:

4,700 = 5,000 + (0.3)(4,000 5,000)

5,090 = 4,700 + (0.3)(6,000 4,700)

4,763 = 5,090 + (0.3)(4,000 5,090)

4,834 = 4,763 + (0.3)(5,000 4,763)

6,384 = 4,834 + (0.3)(10,000 4,834)

6,869 = 6,384 + (0.3)(8,000 6,384)

6,908 = 6,869 + (0.3)(7,000 6,869)

7,536 = 6,908 + (0.3)(9,000 6,908)

8,875 = 7,536 + (0.3)(12,000 7,536)

10,412 = 8,875 + (0.3)(14,000 8,875)

Copyright 2012 Pearson Education, Inc. publishing as Prentice Hall

Forecast Using = 0.6

Forecast Using = 0.9

Total absolute deviation 372.8

MAD=0.3 = 372.8/5 = 74.56

Three-Year Moving Average

(450 + 495 + 518)/3 = 487.67

Copyright 2012 Pearson Education, Inc. publishing as Prentice Hall

(see Problem 5-21)

MADmoving average = 134/2 = 67

Copyright 2012 Pearson Education, Inc. publishing as Prentice Hall

b. The total MAD is 2.60.

Copyright 2012 Pearson Education, Inc. publishing as Prentice Hall