12.

2
Month Demand for Soft Shag Carpet (1000 yd)
1 5
2 10
3 6
4 8
5 14
6 10
7 9
8 12

a. Compute a three-month moving average forecast for months 4 through 9.

Month Demand for Soft Shag Carpet (1000 yd) Forecast Error
1 5 - -
2 10 - -
3 6 - -
4 8 7 1
5 14 8 6
6 10 9.33333333333333 0.666666666667
7 9 10.6666666666667 -1.66666666667
8 12 11 1
9 - 10.3333333333333 -

b. Compute a weighted three-month moving average forecast for months 4 through 9. Assign weights of 0.55, 0.33
Month Demand for Soft Shag Carpet (1000 yd) Forecast Error
1 5 - -
2 10 - -
3 6 - -
4 8 7.2 0.8
5 14 7.58 6.42
6 10 11.06 -1.06
7 9 11.08 -2.08
8 12 9.93 2.07
9 - 10.77 -

c. Compare the two forecasts using MAD. Which forecast appears to be more accurate?
As MAD of 3-month moving average is smaller than that of weighted 3-month moving average: 2.067 < 2.486
3-month moving average is more accurate.

12.3.

Demand for Forecast

Year Quarter
Fertilizer (ton)
 1 1 105
2 150
3 93
4 121
2 5 140
6 170
7 105
8 150
3 9 150
10 170
11 110
12 130

a. Compute a three-quarter moving average forecast for quarters 4 through 13 and compute the forecast error for each quar
Demand for Forecast
Year Quarter
Fertilizer (ton)
1 1 105 -
2 150 -
3 93 -
4 121 116
2 5 140 121.34
6 170 118
7 105 143.67
8 150 138.34
3 9 150 141.67
10 170 135
11 110 156.67
12 130 143.34
1 13 - 136.67
b. Compute a five-quarter moving average forecast for quarters 6 through 13 and compute the forecast error for each quarte
Demand for Forecast
Year Quarter Fertilizer (ton)
1 1 105 -
2 150 -
3 93 -
4 121 -
2 5 140 -
6 170 121.8
7 105 134.8
8 150 125.8
3 9 150 137.2
10 170 143
11 110 149
12 130 137
 1 13 - 142
c. Compute a weighted three-quarter moving average forecast using weights of 0.50, 0.33, and 0.17 for the most recent, next
Demand for Forecast
Year Quarter
Fertilizer (ton)
1 1 105 -
2 150 -
3 93 -
4 121 113.85
2 5 140 116.69
6 170 125.74
7 105 151.77
8 150 132.4
3 9 150 138.55
10 170 142.35
11 110 160
12 130 136.6

d. Compare the forecasts developed in parts (a), (b), and (c) using cumulative error. Which forecast appears to be most accur
3-quarter moving average 31.97
5-quarter moving average 36.4
3-quarter weighted moving average 28.05
3-quarter weighted moving average is more accurate than the others, 3-quarter moving average is more average t
28.05 < 31.97 < 36.4
12.5.
Semester Students Enrolled in OM
1 270
2 310
3 250
4 290
5 370
6 410
7 400
8 450
a. Compute a three-semester moving average forecast for semesters 4 through 9.
Semester Students Enrolled in OM Forecast Error
1 270 - -
2 310 - -
3 250 - -
4 290 276.67 13.33
5 370 283.34 86.66
6 410 303.34 106.66
7 400 356.67 43.33
8 450 393.34 56.66
9 - 420 306.64
b. Compute the exponentially smoothed forecast (α = 0.2) for the enrollment data.
Semester Students Enrolled in OM Forecast Error
1 270 -
2 310 270 40
3 250 278 28
4 290 272.4 17.6
5 370 275.92 94.08
6 410 294.74 115.26
7 400 317.8 82.2
8 450 334.24 115.76
c. Compare the two forecasts using MAD and indicate the most accurate.
492.9 > 306.64
3-semester moving average is more accurate than exponentially smoothed forecast
12.7.
Month Fund Price
1 62.7
2 63.9
3 68
4 66.4
5 67.2
6 65.8
7 68.2
8 69.3
9 67.2
10 70.1
Compute the exponentially smoothed forecast with α = 0.40, the adjusted exponentially smoothed forecast with α = 0.40 an
exponentially smoothed with a=0.4
adjusted exponentially with a=0.4, b=0.3
linear trend line foecast
Exponentially smoothed with a=0.4
Month Fund Price Forecast Error
1 62.7 - -
2 63.9 62.7 1.2
3 68 63.9 4.1
4 66.4 68 -1.6
5 67.2 66.4 0.8
6 65.8 67.2 -1.4
7 68.2 65.8 2.4
8 69.3 68.2 1.1
9 67.2 69.3 -2.1
10 70.1 67.2 2.9
Adjusted exponentially smoothed with a =0.4, b=0.3
Month Fund Price Forecast Trend factor
 1 62.7 - 0
2 63.9 62.7 0
3 68 63.9 0.36
4 66.4 68 1.698
5 67.2 66.4 1.728
6 65.8 67.2 2.487
7 68.2 65.8 2.814
8 69.3 68.2 4.379
9 67.2 69.3 6.023
10 70.1 67.2 7.2
Linear Trend Line
Month Fund Price Linear Trend Error
1 62.7 64.15 -1.45
2 63.9 64.76 -0.86
3 68 65.37 2.63
4 66.4 65.98 0.42
5 67.2 66.59 0.61
6 65.8 67.2 -1.4
7 68.2 67.81 0.39
8 69.3 68.42 0.88
9 67.2 69.03 -1.83
10 70.1 69.64 0.46
sum
Parameter for linear trend line
avg(x) 5.5 cumulative
avg(y) 66.88 mad
b 0.607272727272726
a 63.54
linear trend equation: y = a + bx y = 63.54 + 0.61x

Comparison: Cumulative MAD

Exp Smoothed 7.4 10.93
Adjusted Exp Smoothed -19.289 3.241
Linear -0.15 1.956
linear
exp
 Abs Error

1
6
0.6666666666667
1.6666666666667
1
2.0666666666667

9. Assign weights of 0.55, 0.33, and 0.12 to the months in sequence, starting with the most recent month.
Abs Error

0.8
6.42
1.06
2.08
2.07
2.486

g average: 2.067 < 2.486

Error
e forecast error for each quarter.

Error
-
-
-
5
18.66
52
-38.67
11.66
8.33
35
-46.67
-13.34
31.97
forecast error for each quarter.
Error
-
-
-
-
-
48.2
-29.8
24.2
12.8
27
-39
-7
 36.4
0.17 for the most recent, next recent, and most distant data, respectively, and compute the forecast error for each quarter.
Error
-
-
-
7.15
23.31
44.26
-46.77
17.6
11.45
27.65
-50
-6.6
28.05
ast appears to be most accurate? Do any exhibit any bias?

ing average is more average than that of 5-quarter

 492.9

thed forecast

hed forecast with α = 0.40 and β = 0.30, and the linear trend line forecast. Compare the accuracy of the three forecasts using cumulative e

Abs Error
-
1.2
4.1
1.6
0.8
1.4
2.4
1.1
2.1
7.4 2.9 1.9555555556

Adjusted Forecast Error Abs Error T = b*(F[t+1] - F(t)] + (1+b)T[t]

 - - - AF = F[t+1] + T[t+1]
62.7 1.2 1.2
64.26 3.74 3.74
69.698 -3.298 3.298
68.128 -0.928 0.928
69.687 -3.887 3.887
68.614 -0.414 0.414
72.579 -3.279 3.279
75.323 -8.123 8.123
74.4 -4.3 -19.289 4.3 3.241

x (period) y (demand) xy x^2 abs error

1 62.7 62.7 1 1.45
2 63.9 127.8 4 0.86
3 68 204 9 2.63
4 66.4 265.6 16 0.42
5 67.2 336 25 0.61
6 65.8 394.8 36 1.4
7 68.2 477.4 49 0.39
8 69.3 554.4 64 0.88
9 67.2 604.8 81 1.83
10 70.1 701 100 0.46
55 668.8 3728.5 385

-0.15
10.93
or for each quarter.
three forecasts using cumulative error and MAD, and indicate which forecast appears to be most accurate.