Professional Documents
Culture Documents
Forecasting
Learning Objectives
1.
Understand that the long-run success of an organization is often closely related to how well
management is able to predict future aspects of the operation.
2.
3.
Be able to use smoothing techniques such as moving averages and exponential smoothing.
4.
Be able to use the least squares method to identify the trend component of a time series.
5.
Understand how the classical time series model can be used to explain the pattern or behavior of the
data in a time series and to develop a forecast for the time series.
6.
7.
8.
6-1
Chapter 6
Solutions:
1.
a.
Month
Time-Series
Value
3-Month Moving
Average Forecast
1
2
3
4
5
6
7
8
9
10
11
12
9.5
9.3
9.4
9.6
9.8
9.7
9.8
10.5
9.9
9.7
9.6
9.6
9.40
9.43
9.60
9.70
9.77
10.00
10.07
10.03
9.73
(Error)2
0.04
0.14
0.01
0.01
0.53
0.01
0.14
0.18
0.02
1.08
4-Month Moving
Average Forecast
(Error)2
9.45
9.53
9.63
9.73
9.95
9.98
9.97
9.92
0.12
0.03
0.03
0.59
0.00
0.08
0.14
0.10
1.09
2.
b.
c.
For the limited data provided, the 5-week moving average provides the smallest MSE.
a.
Week
Time-Series
Value
4-Week Moving
Average Forecast
1
2
3
4
5
6
7
8
9
10
11
12
17
21
19
23
18
16
20
18
22
20
15
22
20.00
20.25
19.00
19.25
18.00
19.00
20.00
18.75
b.
(Error)2
4.00
18.06
1.00
1.56
16.00
1.00
25.00
10.56
77.18
6-2
5-Week Moving
Average Forecast
19.60
19.40
19.20
19.00
18.80
19.20
19.00
(Error)2
12.96
0.36
1.44
9.00
1.44
17.64
9.00
51.84
Forecasting
c.
3.
For the limited data provided, the 5-week moving average provides the smallest MSE.
a.
b.
Week
Time-Series
Value
Weighted Moving
Average Forecast
Forecast
Error
1
2
3
4
5
6
7
8
9
10
11
12
17
21
19
23
18
16
20
18
22
20
15
22
19.33
21.33
19.83
17.83
18.33
18.33
20.33
20.33
17.83
3.67
-3.33
-3.83
2.17
-0.33
3.67
-0.33
-5.33
4.17
(Error)2
13.47
11.09
14.67
4.71
0.11
13.47
0.11
28.41
17.39
103.43
c.
You could always find a weighted moving average at least as good as the unweighted one. Actually
the unweighted moving average is a special case of the weighted ones where the weights are equal.
4.
Wee
k
1
2
3
4
5
6
7
8
9
10
11
12
Time-Series
Value
Forecast
Error
(Error)2
17
21
19
23
18
16
20
18
22
20
15
22
17.00
17.40
17.56
18.10
18.09
17.88
18.10
18.09
18.48
18.63
18.27
4.00
1.60
5.44
-0.10
-2.09
2.12
-0.10
3.91
1.52
-3.63
3.73
16.00
2.56
29.59
0.01
4.37
4.49
0.01
15.29
2.31
13.18
13.91
101.72
6-3
Chapter 6
5.
a.
Month
Yt
1
2
3
4
5
6
7
8
9
10
11
12
80
82
84
83
83
84
85
84
82
83
84
83
3-Month Moving
Averages Forecast
82.00
83.00
83.33
83.33
84.00
84.33
83.67
83.00
83.00
(Error)2
1.00
0.00
0.45
2.79
0.00
5.43
0.45
1.00
0.00
11.12
=2
Forecast
80.00
80.40
81.12
81.50
81.80
82.24
82.79
83.03
82.83
82.86
83.09
(Error)2
4.00
12.96
3.53
2.25
4.84
7.62
1.46
1.06
0.03
1.30
0.01
39.06
6.
b.
a.
b.
7.
The more recent data receives the greater weight or importance in determining the forecast. The
moving averages method weights the last n data values equally in determining the forecast.
a.
Month
Time-Series
Value
1
2
3
4
5
6
7
8
9
10
11
12
240
350
230
260
280
320
220
310
240
310
240
230
3-Month Moving
Average Forecast
273.33
280.00
256.67
286.67
273.33
283.33
256.67
286.67
263.33
6-4
(Error)2
= .2
Forecast
(Error)2
177.69
0.00
4010.69
4444.89
1344.69
1877.49
2844.09
2178.09
1110.89
240.00
262.00
255.60
256.48
261.18
272.95
262.36
271.89
265.51
274.41
267.53
12100.00
1024.00
19.36
553.19
3459.79
2803.70
2269.57
1016.97
1979.36
1184.05
1408.50
Forecasting
17,988.52
27,818.49
8.
a.
Day
1
2
3
4
5
6
7
8
9
10
11
12
Time-Series Value
14.45
15.75
16.45
17.40
17.32
15.96
16.45
15.60
15.09
16.42
16.21
15.22
5-Day Moving
Average Forecast
Forecast
Error
(Error)2
16.27
16.58
16.72
16.55
16.08
15.90
15.95
-0.31
-0.13
-1.12
-1.46
0.34
0.31
-0.73
0.10
0.02
1.25
2.12
0.11
0.09
0.54
The weighted moving average forecasts for days 5-12 are 16.49, 17.01, 16.71, 16.57, 16.10, 15.60,
15.09, 16.42, 16.21 and 15.22
Note: MSE = 5.21/8 = 0.65
Forecast for September 4 is 0.1(15.09) + 0.2(16.42) + 0.3(16.21) + 0.4(15.22) = 15.74
c.
The exponential smoothing forecasts for days 2-12 are 14.45, 15.36, 16.12, 17.02, 17.23, 16.34,
16.42, 15.85, 15.32, 16.09 and 16.17
Note: MSE = 9.57/11 = 0.87
Forecast for September 4 is 0.7(15.22) + 0.3(16.17) = 15.51
6-5
Chapter 6
d.
Method
Moving Averages
Weighted Moving Average
Exponential Smoothing
MSE
0.60
0.65
0.87
Moving Averages is the best of the three approaches because it has the smallest MSE.
9.
Note: Results were obtained using the Forecasting module of The Management Scientist.
a.
Method
3-Quarter
4-Quarter
Forecast
80.73
80.55
MSE
2.53
2.81
The 3-quarter moving average forecast is better because it has the smallest MSE.
b.
Method
= .4
= .5
Forecast
80.40
80.57
MSE
2.40
2.01
10. a.
Season
1991-1992
1992-1993
1993-1994
1994-1995
1995-1996
1996-1997
1997-1998
1998-1999
1999-2000
2000-2001
2001-2002
Time-Series
Value
30.1
32.6
29.8
29.3
30.4
29.6
28.7
26.8
29.7
31.1
31.4
Totals:
= 0.1
Forecast
= 0.1
(Error)2
= 0.2
Forecast
= 0.2
(Error)2
30.10
30.35
30.30
30.20
30.22
30.15
30.01
29.69
29.69
29.83
6.2500
0.3025
0.9900
0.0418
0.3794
2.1151
10.2972
0.0001
1.9903
2.4640
24.8304
30.10
30.60
30.44
30.21
30.25
30.12
29.84
29.23
29.32
29.68
6.2500
0.6400
1.2996
0.0353
0.4220
2.0155
9.2157
0.2222
3.1582
2.9642
26.2228
Smoothing Constant
0.1
0.2
MSE
24.8304/10 = 2.48
26.2228/10 = 2.62
6-6
Forecasting
b.
Using = 0.1
Forecast for 2002 - 2003 season = 0.1(31.4) + 0.9(29.83) = 29.99
11. a.
Time Series
Value
28.9
31.0
29.9
30.1
32.2
31.5
32.0
31.9
30.0
Period
1
2
3
4
5
6
7
8
9
= .2
Forecasts
= .3
Forecasts
= .4
Forecasts
29.80
30.04
30.01
30.03
30.46
30.67
30.94
31.13
29.80
30.16
30.08
30.09
30.72
30.95
31.27
31.46
29.80
30.28
30.13
30.12
30.95
31.17
31.50
31.66
b.
12.
t
Yt
Ft
Yt - Ft
(Yt - Ft)2
1
2
3
4
5
6
7
8
9
10
11
12
2,750
3,100
3,250
2,800
2,900
3,050
3,300
3,100
2,950
3,000
3,200
3,150
2,750.00
2,890.00
3,034.00
2,940.40
2,924.24
2,974.54
3,104.73
3,102.84
3,041.70
3,025.02
3,095.01
350.00
360.00
-234.00
-40.40
125.76
325.46
-4.73
-152.84
-41.70
174.98
54.99
Total
122,500.00
129,600.00
54,756.00
1,632.16
15,815.58
105,924.21
22.37
23,260.07
1,738.89
30,618.00
3,023.90
488,991.18
6-7
Chapter 6
13. a & b.
c.
Week
Time-Series
Value
= .2
Forecast
1
2
3
4
5
6
7
8
9
10
7.35
7.40
7.55
7.56
7.60
7.52
7.52
7.70
7.62
7.55
7.35
7.36
7.40
7.43
7.46
7.48
7.48
7.53
7.55
(Error)2
= .3
Forecast
.0025
.0361
.0256
.0289
.0036
.0016
.0484
.0081
.0000
.1548
7.35
7.36
7.42
7.46
7.50
7.51
7.51
7.57
7.58
(Error)2
.0025
.0361
.0196
.0196
.0004
.0001
.0361
.0025
.0009
.1178
14.
The following values are needed to compute the slope and intercept:
t = 21
= 91
= 117.1
tY
= 403.7
Computation of slope:
b1 =
tY d
t Y i / n = 403.7 (21)(117.1) / 6 = 0.3514
91 (21) / 6
t d
ti / n
t
Computation of intercept:
b0 = Y b1 t = 19.5167 (0.3514)(3.5) = 20.7466
A linear trend model is not appropriate. A nonlinear model would provide a better approximation.
16. a.
b.
The following values are needed to compute the slope and intercept:
= 36
= 204
Yt
= 223.8
6-8
t Y t = 1081.6
Forecasting
Computation of slope:
b1 =
tY d
t Y i / n = 10816. (36)(2238. ) / 8 = 17738
.
204 (36) / 8
t d
ti / n
t
Computation of intercept:
b0 = Y - b1 t = 27.975 - 1.7738(4.5) = 19.993
Rating
25
23
21
19
17
15
0
10
Season
t = 105
= 1015
tY
= 301
= 2115.8
Computation of slope:
b1 =
Computation of intercept:
b0 = Y b1 t = 21.5(-0.62286)(7.5) = 26.1714
6-9
15
Chapter 6
65
Percent Voting
60
55
50
45
40
0
Year
The graph shows a linear trend.
b. The following values are needed to compute the slope and intercept:
t = 55
= 385
= 542.77
tY
= 2891.1
Computation of slope:
b1 =
Computation of intercept:
b0 = Y b1 t = (542.77/10) - (-1.141)(55/10) = 60.553
Equation for linear trend: Tt = 60.553 - 1.141t
The average decrease per presidential election is approximately 1.14%
c. 2004 forecast: T8 = 60.553 - 1.141(11) = 48.0%
19. a.
The following values are needed to compute the slope and intercept:
= 78
= 650
Yt
= 343
t Y t = 2441
Computation of slope:
b1 =
tY d
t Y i / n = 2441 (78)(343) / 12 = 1479
.
650 (78) / 12
/
t
t
n
d
i
t
6 - 10
10
12
Forecasting
Computation of intercept:
b0 = Y - b1 t = (343/12) - 1.479(78/12) = 18.97
Expenditures
40
35
30
25
20
15
10
0
10
12
14
Time
c.
20. a.
b.
= 15
= 55
Yt
= 200
t Y t = 750
Computation of slope:
b1 =
tY d
t Y i / n = 750 (15)(200) / 5 = 15
55 (15) / 5
t d
ti / n
t
Computation of intercept:
b0 = Y - b1 t = 40 - 15(3) = -5
6 - 11
Chapter 6
21. a.
b.
= 28
= 140
Yt
= 595
t Y t = 2815
Computation of slope:
b1 =
tY d
t Y i / n = 2815 (28)(595) / 7 = 155357
.
140 (28) / 7
t d
ti / n
t
Computation of intercept:
b0 = Y - b1 t = 85 - 15.5357(4) = 22.857
b.
T2 = 6.4564 + 0.5345t
c.
5.345 million
d.
23.
Note: Results were obtained using the Forecasting module of The Management Scientist.
a.
Smoothing Constant
= .3
= .4
= .5
MSE
4,492.37
2,964.67
2,160.31
Tt = 244.778 + 22.088t
MSE = 357.81
c.
Trend projection provides much better forecasts because it has the smallest MSE. The reason MSE
is smaller for trend projection is that sales are increasing over time; as a result, exponential
smoothing continuously underestimates the value of sales. If you look at the forecast errors for
exponential smoothing you will see that the forecast errors are positive for periods 2 through 18.
6 - 12
Forecasting
24.
Note: Results were obtained using the forecasting module of The Management Scientist.
a.
b.
Tt = 149.719 + 18.451t
Forecast for July is 278.88
Forecast for August is 297.33
c.
25. a.
The proposed settlement is not fair since it does not account for the upward trend in sales. Based
upon trend projection, the settlement should be based on forecasted lost sales of $278,880 in July
and $297,330 in August.
Four quarter moving averages beginning with
(1690 + 940 + 2625 + 2500) / 4 = 1938.75
Other moving averages are
1966.25
1956.25
2025.00
1990.00
2002.50
2052.50
2060.00
2123.75
b.
Quarter
1
2
3
4
Seasonal-Irregular
Component Values
0.904
0.900
0.448
0.526
1.344
1.453
1.275
1.164
Seasonal Index
0.9020
0.4970
1.3985
1.2195
4.0070
Adjusted Seasonal
Index
0.900
0.486
1.396
1.217
The largest seasonal effect is in the third quarter which corresponds to the back-to-school
demand during July, August, and September of each year.
6 - 13
Chapter 6
26.
Seasonal-Irregular
Component Values
0.72
0.70
0.80
0.75
0.83
0.82
0.94
0.99
1.01
1.02
1.25
1.36
1.49
1.51
1.19
1.26
0.98
0.97
0.98
1.00
0.93
0.94
0.78
0.80
Month
1
2
3
4
5
6
7
8
9
10
11
12
Seasonal Index
0.71
0.78
0.83
0.97
1.02
1.31
1.50
1.23
0.98
0.99
0.94
0.79
12.05
Use a twelve period moving averages. After centering the moving averages, you should obtain the
following seasonal indexes:
Hour
1
2
3
4
5
6
b.
Seasonal Index
0.771
0.864
0.954
1.392
1.571
1.667
Hour
7
8
9
10
11
12
Seasonal Index
1.207
0.994
0.850
0.647
0.579
0.504
The hours of July 18 are number 37 to 48 in the time series. Thus the trend component for 7:00 a.m.
on July 18 (period 37) would be
T37 = 32.983 + .3922(37) = 47.49
A summary of the trend components for the twelve hours on July 18 is as follows:
Hour
1
2
3
4
5
6
c.
Trend Component
47.49
47.89
48.28
48.67
49.06
49.46
Hour
7
8
9
10
11
12
Trend Component
49.85
50.24
50.63
51.02
51.42
51.81
Multiply the trend component in part b by the seasonal indexes in part a to obtain the twelve hourly
forecasts for July 18. For example, 47.49 x (.771) = 36.6 or rounded to 37, would be the forecast for
7:00 a.m. on July 18th.
6 - 14
Forecasting
Forecast
37
41
46
68
77
82
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
26
27
28
Sales
6
15
10
4
10
18
15
7
14
26
23
12
19
28
25
18
22
34
28
21
24
36
30
20
28
40
35
27
Hour
7
8
9
10
11
12
Forecast
60
50
43
33
30
26
Centered
Moving Average
Seasonal-Irregular
Component
9.250
10.125
11.125
12.125
13.000
14.500
16.500
18.125
19.375
20.250
20.750
21.750
22.875
24.000
25.125
25.875
26.500
27.000
27.500
27.625
28.000
29.000
30.125
31.625
1.081
0.395
0.899
1.485
1.154
0.483
0.848
1.434
1.187
0.593
0.916
1.287
1.093
0.750
0.876
1.314
1.057
0.778
0.873
1.303
1.071
0.690
0.929
1.265
28. a.
b.
Quarter
1
2
3
4
Seasonal-Irregular
Component Values
0.899, 0.848, 0.916, 0.876, 0.873, 0.929
1.485, 1.434, 1.287, 1.314, 1.303, 1.265
1.081, 1.154, 1.187, 1.093, 1.057, 1.071
0.395, 0.483, 0.593, 0.750, 0.778, 0.690
Total
6 - 15
Seasonal Index
0.890
1.348
1.107
0.615
3.960
Chapter 6
Quarter
1
2
3
4
Hudson Marine experiences the largest seasonal increase in quarter 2. Since this quarter occurs
prior to the peak summer boating season, this result seems reasonable.
29. a.
t
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Sales
4
2
1
5
6
4
4
14
10
3
5
16
12
9
7
22
18
10
13
35
Quarter
Centered
Moving Average
Seasonal-Irregular
Component
3.250
3.750
4.375
5.875
7.500
7.875
7.875
8.250
8.750
9.750
10.750
11.750
13.250
14.125
15.000
17.375
0.308
1.333
1.371
0.681
0.533
1.778
1.270
0.364
0.571
1.641
1.116
0.766
0.528
1.558
1.200
0.576
Seasonal-Irregular
Component Values
Seasonal
Index
1
2
3
4
Quarter
1
2
3
4
Adjusted
Seasonal Index
1.271
0.613
0.498
1.619
6 - 16
1.239
0.597
0.485
1.578
3.899
Forecasting
b.
30. a.
Note: To simplify the calculations the seasonal indexes calculated in problem 28 have been rounded
to two decimal places.
Year
Quarter
Sales Yt
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
6
15
10
4
10
18
15
7
14
26
23
12
19
28
25
18
22
34
28
21
24
36
30
20
28
40
35
27
Seasonal Factor
St
0.90
1.36
1.12
0.62
0.90
1.36
1.12
0.62
0.90
1.36
1.12
0.62
0.90
1.36
1.12
0.62
0.90
1.36
1.12
0.62
0.90
1.36
1.12
0.62
0.90
1.36
1.12
0.62
6 - 17
Deseasonalized Sales
Yt / St = TtIt
6.67
11.03
8.93
6.45
11.11
13.24
13.39
11.29
15.56
19.12
20.54
19.35
21.11
20.59
22.32
29.03
24.44
25.00
25.00
33.87
26.67
26.47
26.79
32.26
31.11
29.41
31.25
43.55
Chapter 6
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
26
27
28
406
t = 14.5
Yt
(deseasonalized)
6.67
11.03
8.93
6.45
11.11
13.24
13.39
11.29
15.56
19.12
20.54
19.35
21.11
20.59
22.32
29.03
24.44
25.00
25.00
33.87
26.67
26.47
26.79
32.26
31.11
29.41
31.25
43.55
605.55
Y = 21.627
b1 = 1.055
tYt
t2
1
4
9
16
25
36
49
64
81
100
121
144
169
196
225
256
289
324
361
400
441
484
529
576
625
676
729
784
7,714
6.67
22.06
26.79
25.80
55.55
79.44
93.73
90.32
140.04
191.20
225.94
232.20
274.43
288.26
334.80
464.48
415.48
450.00
475.00
677.40
560.07
582.34
616.17
774.24
777.75
764.66
843.75
1,219.40
10,707.34
b0 = 6.329
T t = 6.329 + 1.055t
b.
t
29
30
31
32
Trend Forecast
36.92
37.98
39.03
40.09
c.
Year
8
Quarter
1
2
3
4
Trend Forecast
36.92
37.98
29.03
40.09
6 - 18
Seasonal Index
0.90
1.36
1.12
0.62
Quarterly Forecast
33.23
51.65
43.71
24.86
Forecasting
31. a.
Note: To simplify the calculations the seasonal indexes in problem 29 have been round to two
decimal places.
Year
Quarter
Sales Yt
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
4
2
1
5
6
4
4
14
10
3
5
16
12
9
7
22
18
10
13
35
t
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
210
Seasonal Factor
St
1.27
0.61
0.50
1.62
1.27
0.61
0.50
1.62
1.27
0.61
0.50
1.62
1.27
0.61
0.50
1.62
1.27
0.61
0.50
1.62
Yt
(deseasonalized)
3.15
3.28
2.00
3.09
4.72
6.56
8.00
8.64
7.87
4.92
10.00
9.88
9.45
14.75
14.00
13.58
14.17
16.39
26.00
21.60
202.05
6 - 19
tYt
3.15
6.56
6.00
12.36
23.60
39.36
56.00
69.12
70.83
49.20
110.00
118.56
122.85
206.50
210.00
217.28
240.89
295.02
494.00
432.00
2783.28
Deseasonalized Sales
Yt / St = TtIt
3.15
3.28
2.00
3.09
4.72
6.56
8.00
8.64
7.87
4.92
10.00
9.88
9.45
14.75
14.00
13.58
14.17
16.39
26.00
21.60
t2
1
4
9
16
25
36
49
64
81
100
121
144
169
196
225
256
289
324
361
400
2870
Chapter 6
t = 10.5
Y = 10.1025
b1 = .995
b0 = - .345
T t = - .345 + .995 t
b.
y
21
22
23
24
Trend Forecast
20.55
21.55
22.54
23.54
c.
Year
6
32.
Trend
Forecast
20.55
21.55
22.54
23.54
Quarter
1
2
3
4
Seasonal
Index
1.27
0.61
0.50
1.62
Quarterly
Forecast
26.10
13.15
11.27
38.13
Note: Results were obtained using the Forecasting module of The Management Scientist.
a.
Seasonal Index
1.696
1.458
0.711
0.326
0.448
1.362
Time Period
12 - 4 p.m.
4 - 8 p.m.
Forecast
166,761.13
146,052.99
b.
33.
x=7
b.
Restaurant
(i)
xi
yi
xi yi
xi2
1
2
3
4
5
Totals
1
4
6
10
14
35
19
44
40
52
53
208
19
176
240
520
742
1,697
1
16
36
100
196
349
y = 41.6
b1 = 2.317
b0 = 25.381
6 - 20
y = 25.381 + 2.317 x
Forecasting
34.
x=3
b.
xi
yi
xiyi
1
1
2
3
3
4
5
5
24
36
33
31
29
27
25
23
20
224
36
33
62
87
81
100
115
100
614
y = 28
b1 = 3.222
b0 = 37.666
xi2
1
1
4
9
9
16
25
25
90
y = 37.666 3.222 x
35.
x = 35
b.
xi
yi
xiyi
20
20
40
30
60
40
210
21
19
15
16
14
17
102
420
380
600
480
840
680
3400
y = 17
b1 = 0.1478
b0 = 22.713
xi2
400
400
1600
900
3600
1600
8500
y = 22.713 0.1478 x
6 - 21