4.

2
Year 1 2 3 4 5 6 7 8 9 10 11
Demand 7 9 5 9 13 8 12 13 9 11 7

a. Starting in year 4 and going to year 12, forecast demand using a 3-year moving average.
b. Starting in year 4 and going to year 12, forecast demand using a 3-year moving average with
weights of .1, .3, and .6; using .6 for the most recent year.
4.2
Year 1 2 3 4 5 6 7 8 9 10 11
Demand 7 9 5 9 13 8 12 13 9 11 7
a. Starting in year 4 and going to year 12, forecast demand using a 3-year moving average.

Year Demand a. Forecast

1 7 n/a
2 9 n/a
3 5 n/a
4 9 =(7+9+5)/3 7.00
5 13 =(9+5+9)/3 7.67
6 8 =(5+9+13)/3 9.00
7 12 =(9+13+8)/3 10.00
8 13 =(13+8+12)/3 11.00
9 9 =(8+12+13)/3 11.00
10 11 =(12+13+9)/3 11.33
11 7 =(13+9+11)/3 11.00
12 =(9+11+7)/3 9.00
4.2
Year 1 2 3 4 5 6 7 8 9 10 11
Demand 7 9 5 9 13 8 12 13 9 11 7
a. Starting in year 4 and going to year 12, forecast demand using a 3-year moving average.
b. Starting in year 4 and going to year 12, forecast demand using a 3-year moving average with weights
of .1, .3, and .6; using .6 for the most recent year.

Year Demand a. Forecast b. Forecast

1 7 n/a n/a
2 9 n/a n/a
3 5 n/a n/a
4 9 =(7+9+5)/3 7.00 =(.1*7 + .3*9 + .6*5)/1 6.40
5 13 =(9+5+9)/3 7.67 =(.1*9 + .3*5 + .6*9)/1 7.80
6 8 =(5+9+13)/3 9.00 =(.1*5+ .3*9 + .6*13)/1 11.00
7 12 =(9+13+8)/3 10.00 =(.1*9 + .3*13 + .6*8)/1 9.60
8 13 =(13+8+12)/3 11.00 =(.1*13 + .3*8 + .6*12)/1 10.90
9 9 =(8+12+13)/3 11.00 =(.1*8 + .3*12 + .6*13)/1 12.20
10 11 =(12+13+9)/3 11.33 =(.1*12 + .3*13 + .6*9)/1 10.50
11 7 =(13+9+11)/3 11.00 =(.1*13 + .3*9 + .6*11)/1 10.60
12 =(9+11+7)/3 9.00 =(.1*9 + .3*11 + .6*7)/1 8.40
4.6 The monthly sales for Yazici Batteries, Inc., were as follows:

Month Sales a. Forecast January sales using each of the following:

Jan 20 1. Naïve method
Feb 21 2. A 3-month moving average
Mar 15 3. A 6-month weighted average using .1, .1, .1, .2, .2, and .3, with the heaviest weights app
Apr 14 4. Exponential smoothing using an α=0.3 and a September forecast of 18
May 13 5. Trend projection
Jun 16 b. With the data given, which method would allow you to forecast next March's sales?
Jul 17
Aug 18
Sep 20
Oct 20
Nov 21
Dec 23
with the heaviest weights applied to the most recent months
recast of 18

ast next March's sales?

4.6 The monthly sales for Yazici Batteries, Inc., were as follows: a. Forecast January sales using each of the fo
1. Naïve method = last month's forecast

Month Sales
Jan 20
Feb 21
Mar 15
Apr 14
May 13
Jun 16
Jul 17
Aug 18
Sep 20
Oct 20
Nov 21
Dec 23
ales using each of the following:
= last month's forecast = 23.00
4.6 The monthly sales for Yazici Batteries, Inc., were as follows: a. Forecast January sales using each of the fo
1. Naïve method = last month's forecast

Month Sales 2. A 3-month moving average

Jan 20
Feb 21
Mar 15
Apr 14
May 13
Jun 16
Jul 17
Aug 18
Sep 20
Oct 20
Nov 21
Dec 23
ales using each of the following:
= last month's forecast = 23.00

ing average =(20 + 21 + 23) / 3 = 21.33

4.6 The monthly sales for Yazici Batteries, Inc., were as follows: a. Forecast January sales using each of the fo
1. Naïve method = last month's forecast

Month Sales 2. A 3-month moving average

Jan 20
Feb 21 3. A 6-month weighted average using .1, .1
Mar 15 heaviest weights applied to the most rec
Apr 14
May 13
Jun 16
Jul 17
Aug 18
Sep 20
Oct 20
Nov 21
Dec 23
ales using each of the following:
= last month's forecast = 23.00

ing average =(20 + 21 + 23) / 3 = 21.33

ghted average using .1, .1, .1, .2, .2, and .3, with the
s applied to the most recent months
= (.1*17 + .1*18 + .1*20 + .2*20 + .2*21 + .3*23) / 1 = 20.60
 Month Sales Error α * error Forecast Ft = Ft – 1 + α(At – 1 - Ft – 1)
Jan 20
Feb 21 4. Exponential smoothing
Mar 15
Apr 14
May 13
Jun 16
Jul 17
Aug 18
Sep 20 18.00
Oct 20
Nov 21
Dec 23
Jan
Ft = Ft – 1 + α(At – 1 - Ft – 1)

4. Exponential smoothing using an α=0.3 and a September forecast of 18

4.6 The monthly sales for Yazici Batteries, Inc., were as follows: a. Forecast January sales using each of the fo
1. Naïve method = last month's forecast
α (At-1 - Ft-1) Ft
Month Sales Error α * error Forecast 2. A 3-month moving average
Jan 20
Feb 21 3. A 6-month weighted average using .1, .
Mar 15 heaviest weights applied to the most re
Apr 14
May 13
Jun 16 4. Exponential smoothing using an α=0.3 a
Jul 17 See table to the left
Aug 18
Sep 20 2.00 0.60 18.00
Oct 20 1.40 0.42 18.60
Nov 21 1.98 0.59 19.02
Dec 23 3.39 1.02 19.61
Jan 20.63
ales using each of the following:
= last month's forecast = 23.00

ing average =(20 + 21 + 23) / 3 = 21.33

ghted average using .1, .1, .1, .2, .2, and .3, with the
s applied to the most recent months
= (.1*17 + .1*18 + .1*20 + .2*20 + .2*21 + .3*23) / 1 = 20.60

oothing using an α=0.3 and a September forecast of 18

Ft = Ft – 1 + α(At – 1 - Ft – 1) 20.63
4.6 The monthly sales for Yazici Batteries, Inc., were as follows: a. Forecast January sales using each of the fo
1. Naïve method = last month's forecast
x y
Month Sales x2 xy 2. A 3-month moving average
Jan 1 20 1.00 20.00
Feb 2 21 4.00 42.00 3. A 6-month weighted average using .1, .
Mar 3 15 9.00 45.00 heaviest weights applied to the most re
Apr 4 14 16.00 56.00
May 5 13 25.00 65.00
Jun 6 16 36.00 96.00 4. Exponential smoothing using an α=0.3 a
Jul 7 17 49.00 119.00
Aug 8 18 64.00 144.00
Sep 9 20 81.00 180.00 5. Trend projection
Oct 10 20 100.00 200.00 See table to the left
Nov 11 21 121.00 231.00
Dec 12 23 144.00 276.00

Sum 78 218 650 1474

Average 6.50 18.17
ales using each of the following:
= last month's forecast = 23.00

ing average =(20 + 21 + 23) / 3 = 21.33

ghted average using .1, .1, .1, .2, .2, and .3, with the
s applied to the most recent months
= (.1*17 + .1*18 + .1*20 + .2*20 + .2*21 + .3*23) / 1 = 20.60

oothing using an α=0.3 and a September forecast of 18

Ft = Ft – 1 + α(At – 1 - Ft – 1) 20.63

20.75
b = (1,474 - 12 * 6.5 * 18.17) / (650 - 12 * 6.5 2)
= 0.396783

a = 18.17 - .3986 * 6.5

= 15.59091

y = 15.59 + .3967 x => 15.59 + .3967 * 13

= 20.74909
4.6 The monthly sales for Yazici Batteries, Inc., were as follows: a. Forecast January sales using each of the fo
1. Naïve method = last month's forecast
x y
Month Sales x2 xy 2. A 3-month moving average
1 20 1.00 20.00
2 21 4.00 42.00 3. A 6-month weighted average using .1, .
3 15 9.00 45.00 heaviest weights applied to the most re
4 14 16.00 56.00
5 13 25.00 65.00
6 16 36.00 96.00 4. Exponential smoothing using an α=0.3 a
7 17 49.00 119.00
8 18 64.00 144.00
9 20 81.00 180.00 5. Trend projection
10 20 100.00 200.00 See table to the left
11 21 121.00 231.00
12 23 144.00 276.00

Sum 78 218 650 1474

Average 6.50 18.17
ales using each of the following:
= last month's forecast = 23.00

ing average =(20 + 21 + 23) / 3 = 21.33

ghted average using .1, .1, .1, .2, .2, and .3, with the
s applied to the most recent months
= (.1*17 + .1*18 + .1*20 + .2*20 + .2*21 + .3*23) / 1 = 20.60

oothing using an α=0.3 and a September forecast of 18

Ft = Ft – 1 + α(At – 1 - Ft – 1) 20.63

20.76
b = (1,474 - 12 * 6.5 * 18.17) / (650 - 12 * 6.5 2)
= 0.398601

b. With the data given, which method

a = 18.17 - .3986 * 6.5 would allow you to forecast next March's
= 15.5791 sales?

y = 15.579 + .3986 x => 15.579 + .3986 * 13

= 20.7608
4.8 Daily high temperatures in St. Louis for the last week were as follows:
93, 94, 93, 95, 96, 88, 90 (yesterday)

a. Forecast the high temperature today, using a 3-day moving average.

b. Forecast the high temperature today, using a 2-day moving average.
c. Calculate the mean absolute deviation (MAD) based on a 2-day moving average.
d. Compute the mean squeared error (MSE) for the 2-day moving average.
e. Calculate the mean absolute percent error (MAPE) for the 2-day moving average.
4.8 Daily high temperatures in St. Louis for the last week were as follows: a. Forecast the high temp
93, 94, 93, 95, 96, 88, 90 (yesterday)

Day Temp
1 93
2 94
3 93
4 95
5 96
6 88
7 90
Today
a. Forecast the high temperature today, using a 3-day moving average.
=(96 + 88 + 90) / 3 = 91.33
4.8 Daily high temperatures in St. Louis for the last week were as follows: a. Forecast the high temp
93, 94, 93, 95, 96, 88, 90 (yesterday)

Day Temp b. Forecast the high temp

1 93
2 94
3 93
4 95
5 96
6 88
7 90
Today
a. Forecast the high temperature today, using a 3-day moving average.
=(96 + 88 + 90) / 3 = 91.33

b. Forecast the high temperature today, using a 2-day moving average.

=(88 + 90) / 2 = 89.00
4.8 Daily high temperatures in St. Louis for the last week were as follows: a. Forecast the high temp
93, 94, 93, 95, 96, 88, 90 (yesterday)

Day Temp 2 day Avg |Diff| b. Forecast the high temp

1 93
2 94
3 93 93.50 0.50 c. Calculate the mean abs
4 95 93.50 1.50
5 96 94.00 2.00
6 88 95.50 7.50
7 90 92.00 2.00
Sum 13.50
a. Forecast the high temperature today, using a 3-day moving average.
=(96 + 88 + 90) / 3 = 91.33

b. Forecast the high temperature today, using a 2-day moving average.

=(88 + 90) / 2 = 89.00

c. Calculate the mean absolute deviation (MAD) based on a 2-day moving average.
MAD = Σ|Actual - Forecast| / n
= 13.50 / 5 2.70
4.8 Daily high temperatures in St. Louis for the last week were as follows: a. Forecast the high temp
93, 94, 93, 95, 96, 88, 90 (yesterday)

Day Temp 2 day Avg |Diff| error2 b. Forecast the high temp
1 93
2 94
3 93 93.50 0.50 0.25 c. Calculate the mean abs
4 95 93.50 1.50 2.25
5 96 94.00 2.00 4.00
6 88 95.50 7.50 56.25
7 90 92.00 2.00 4.00 d. Compute the mean squ
Sum 13.50 66.75
a. Forecast the high temperature today, using a 3-day moving average.
=(96 + 88 + 90) / 3 = 91.33

b. Forecast the high temperature today, using a 2-day moving average.

=(88 + 90) / 2 = 89.00

c. Calculate the mean absolute deviation (MAD) based on a 2-day moving average.
MAD = Σ|Actual - Forecast| / n
= 13.50 / 5 2.70

d. Compute the mean squeared error (MSE) for the 2-day moving average.
MSE = Σ(forecast errors)2 / n
= 66.75 / 5 13.35
4.8 Daily high temperatures in St. Louis for the last week were as follows: a. Forecast the high temp
93, 94, 93, 95, 96, 88, 90 (yesterday)
100*|Diff|
Day Temp 2 day Avg |Diff| error2 / Actual b. Forecast the high temp
1 93
2 94
3 93 93.50 0.50 0.25 0.54 c. Calculate the mean abs
4 95 93.50 1.50 2.25 1.58
5 96 94.00 2.00 4.00 2.08
6 88 95.50 7.50 56.25 8.52
7 90 92.00 2.00 4.00 2.22 d. Compute the mean squ
Sum 13.50 66.75 14.94

e. Calculate the mean abs

a. Forecast the high temperature today, using a 3-day moving average.
=(96 + 88 + 90) / 3 = 91.33

b. Forecast the high temperature today, using a 2-day moving average.

=(88 + 90) / 2 = 89.00

c. Calculate the mean absolute deviation (MAD) based on a 2-day moving average.
MAD = Σ|Actual - Forecast| / n
= 13.50 / 5 2.70

d. Compute the mean squeared error (MSE) for the 2-day moving average.
MSE = Σ(forecast errors)2 / n
= 66.75 / 5 13.35

e. Calculate the mean absolute percent error (MAPE) for the 2-day moving average.
MAPE = (Σ100|Actual - Forecast| / Actual) / n
= 14.94 / 5 2.99 %
Year 1 2 3 4 5 6 7 8 9 10 11
Registrations 4 6 4 5 10 8 7 9 12 14 15

a. Develop a 3-year moving average to forecast registrations from year 4 to year 12

b. Estimate demand again for years 4 to 12 with a 3-year weighted moving average in
which registrations in the most recent year are given a weight of 2, and registrations in the
other 2 years are each given a weight of 1.
 a. Develop a 3-year mo

Year 1 2 3 4 5 6 7 8 9 10 11
Registrations 4 6 4 5 10 8 7 9 12 14 15 Year
1
2
3
4
5
6
7
8
9
10
11
12
a. Develop a 3-year moving average to forecast registrations from year 4 to year 12

Registrations a. Forecast
4
6
4
5 4.67 =(4+6+4)/3
10 5.00 =(6+4+5)/3
8 6.33 =(4+5+10)/3
7 7.67 =(5+10+8)/3
9 8.33 =(10+8+7)/3
12 8.00 =(8+7+9)/3
14 9.33 =(7+9+12)/3
15 11.67 =(9+12+14)/3
13.67 =(12+14+15)/3
 a. Develop a 3-year mo
b. Estimate demand aga
most recent year are gi
Year 1 2 3 4 5 6 7 8 9 10 11
Registrations 4 6 4 5 10 8 7 9 12 14 15 Year
1
2
3
4
5
6
7
8
9
10
11
12
a. Develop a 3-year moving average to forecast registrations from year 4 to year 12
b. Estimate demand again for years 4 to 12 with a 3-year weighted moving average in which registrations in the
most recent year are given a weight of 2, and registrations in the other 2 years are each given a weight of 1.

Registrations a. Forecast b. Forecast

4
6
4
5 4.67 4.50 =(1*4 + 1*6 + 2*4) / 4
10 5.00 5.00 =(1*6 + 1*4 + 2*5) / 4
8 6.33 7.25 =(1*4 + 1*5 + 2*10) / 4
7 7.67 7.75 =(1*5 + 1*10 + 2*8) / 4
9 8.33 8.00 =(1*10 + 1*8 + 2*7) / 4
12 8.00 8.25 =(1*8 + 1*7 + 2*9) / 4
14 9.33 10.00 =(1*7 + 1*9 + 2*12) / 4
15 11.67 12.25 =(1*9 + 1*12 + 2*14) / 4
13.67 14.00 =(1*12 + 1*14 + 2*15) / 4
Forecast = F1 + a( F1 - F2)

Year Registrations Forecast Error Absolute

1 4,000 5,000.00 -1,000.00 1,000.00
2 6,000 4,700.00 1,300.00 1,300.00
3 4,000 5,090.00 -1,090.00 1,090.00
4 5,000 4,763.00 237.00 237.00
5 10,000 4,834.10 5,165.90 5,165.90
6 8,000 6,383.87 1,616.13 1,616.13
7 7,000 6,868.71 131.29 131.29
8 9,000 6,908.10 2,091.90 2,091.90
9 12,000 7,535.67 4,464.33 4,464.33
10 14,000 8,874.97 5,125.03 5,125.03
11 15,000 10,412.48 4,587.52 4,587.52
12
Total 22,629.10 26,809.10
Average 2,057.19 2,437.19
BIAS MAD
 Squared
1,000,000.00
1,690,000.00
1,188,100.00
56,169.00
26,686,522.81
2,611,876.18
17,237.06
4,376,045.61
19,930,242.35
26,265,932.50
21,045,339.75

104,867,465.26
9,533,405.93
MSE
 0.1
Week Actual Sales Naive Method Exponential Smoothing
1 95
2 110 95 95
3 120 110 96.5
4 98 120 98.85
5 112 98 98.77
6 95 112 100.09
7 105 95 99.58
8 130 105 100.12
9 70 130 103.11
10 80 70 99.8
11 110 80 97.82
12 110 99.04
 Deviation

15
23.5
0.85
13.23
5.09
5.42
29.88
33.11
19.8
12.18
158.06

a=0.1 '= 15.81

4.20 Income at the architectural firm Spraggins and Yunes for the period February
to July was as follows:
Month Feb Mar Apr May Jun Jul
Income 70 68.5 64.8 71.7 71.3 72.8

a. Use Trend-adjusted exponential smoothing to forecast the firm's August income.

Assume that the initial forecast average for February is $65,000 and the initial trend
adjustment is 0. The smoothing constants selected are α=.1 and β=.8

b. Assume the smoothing constants selected are α=.1 and β=.3

c. Using MSE, determine which smoothing constants provide a better forecast.

4.20 Income at the architectural firm Spraggins and Yunes for the period February a. Use Trend-adjusted expon
to July was as follows: Assume that the initial foreca
Month Feb Mar Apr May Jun Jul adjustment is 0. The smooth
Income 70 68.5 64.8 71.7 71.3 72.8

Month
Feb
Mar
Apr
May
Jun
Jul
Aug
a. Use Trend-adjusted exponential smoothing to forecast the firm's August income.
Assume that the initial forecast average for February is $65,000 and the initial trend
adjustment is 0. The smoothing constants selected are α=.1 and β=.8

Ft = α(At - 1) + (1 - α)(Ft - 1 + Tt - 1)

Income Ft Tt Forecast
70.00 65.00 0.00 65.00 Tt = b(Ft - Ft - 1) + (1 - b)Tt - 1
68.50 65.50 0.40 65.90
64.80 α= 0.1
71.70 b= 0.8
71.30
72.80
4.20 Income at the architectural firm Spraggins and Yunes for the period February to
July was as follows:
Month Feb Mar Apr May Jun Jul
Income 70 68.5 64.8 71.7 71.3 72.8

Month Income Ft
Feb 70.00 65.00
Mar 68.50 65.50
Apr 64.80 66.16
May 71.70 66.57
Jun 71.30 67.49
Jul 72.80 68.61
Aug 69.99
Ft = α(At - 1) + (1 - α)(F

Tt = b(Ft - Ft - 1) + (1 - b)Tt - 1
 a. Use Trend-adjusted exponential smoothing to forecast the firm's August income.
Assume that the initial forecast average for February is $65,000 and the initial trend
adjustment is 0. The smoothing constants selected are α=.1 and β=.8

Smooth
Ft Tt a. Forecast
0.00 65.00
=.1*(70) + .9(65 + 0) 0.40 =.8(65.5 - 65) + .8(0) 65.90 = 65.5 + 0.4
=.1*(68.5) + .9(65.5 + 0.4) 0.61 =.8(66.16 - 65.5) + .8(0.4) 66.77 = 66.16 + 0.61
=.1*(64.8) + .9(66.16 + 0.61) 0.45 =.8(66.57 - 66.16) + .8(0.61) 67.02 = 66.57 + 0.45
=.1*(71.7) + .9(66.57 + 0.45) 0.82 =.8(67.49 - 66.57) + .8(0.45) 68.31 = 67.49 + 0.82
=.1*(71.3) + .9(67.49 + 0.82) 1.06 =.8(68.61 - 67.49) + .8(0.82) 69.68 = 68.61 + 1.06
=.1*(72.8) + .9(68.61 + 1.06) 1.31 =.8(69.99 - 68.61) + .8(1.06) 71.30 = 69.99 + 1.31
t = α(At - 1) + (1 - α)(Ft - 1 + Tt - 1)

t = b(Ft - Ft - 1) + (1 - b)Tt - 1
4.20 Income at the architectural firm Spraggins and Yunes for the period February a. Use Trend-adjusted expon
to July was as follows: Assume that the initial foreca
Month Feb Mar Apr May Jun Jul adjustment is 0. The smooth
Income 70 68.5 64.8 71.7 71.3 72.8
b. Assume the smoothing con

Month
Feb
Mar
Apr
May
Jun
Jul
Aug
a. Use Trend-adjusted exponential smoothing to forecast the firm's August income.
Assume that the initial forecast average for February is $65,000 and the initial trend
adjustment is 0. The smoothing constants selected are α=.1 and β=.8

b. Assume the smoothing constants selected are α=.1 and β=.3

Income Ft Tt Forecast
Ft = α(At - 1) + (1 - α)(Ft - 1 + Tt - 1)
70.00 65.00 0.00 65.00
68.50
64.80 Tt = b(Ft - Ft - 1) + (1 - b)Tt - 1
71.70
71.30
72.80
4.20 Income at the architectural firm Spraggins and Yunes for the period February to a. Use Trend-adjusted ex
July was as follows: Assume that the initial fo
Month Feb Mar Apr May Jun Jul adjustment is 0. The smo
Income 70 68.5 64.8 71.7 71.3 72.8
b. Assume the smoothin

Month Income a. Forecast Ft

Feb 70.00 65.00 65.00
Mar 68.50 65.90 65.50
Apr 64.80 66.77 65.94
May 71.70 67.02 66.03
Jun 71.30 68.31 66.78
Jul 72.80 69.68 67.55
Aug 71.30 68.51
Ft = α(At - 1) + (1 - α)(F

Tt = b(Ft - Ft - 1) + (1 - b)Tt
a. Use Trend-adjusted exponential smoothing to forecast the firm's August income.
Assume that the initial forecast average for February is $65,000 and the initial trend
adjustment is 0. The smoothing constants selected are α=.1 and β=.8

b. Assume the smoothing constants selected are α=.1 and β=.3

Smooth
Ft Tt b. Forecast
0.00 65.00
=.1*(70) + .9(65 + 0) 0.15 =.3(65.5 - 65) + .7(0) 65.65 = 65.5 + 0.15
=.1*(68.5) + .9(65.5 + 0.15) 0.24 =.3(65.94 - 65.5) + .7(0.15) 66.17 = 65.94 + 0.24
=.1*(64.8) + .9(65.94 + 0.24) 0.19 =.3(66.03 - 65.94) + .7(0.24) 66.23 = 66.03 + 0.19
=.1*(71.7) + .9(66.03 + 0.19) 0.36 =.3(66.78 - 66.03) + .7(0.19) 67.13 = 66.78 + 0.36
=.1*(71.3) + .9(66.78 + 0.36) 0.48 =.3(67.55 - 66.78) + .7(0.36) 68.03 = 67.55 + 0.48
=.1*(72.8) + .9(67.55 + 0.48) 0.63 =.3(68.51 - 67.55) + .7(0.48) 69.14 = 68.51 + 0.63
Ft = α(At - 1) + (1 - α)(Ft - 1 + Tt - 1)

Tt = b(Ft - Ft - 1) + (1 - b)Tt - 1
4.20 Income at the architectural firm Spraggins and Yunes for the period February to a. Use Trend-adjusted ex
July was as follows: Assume that the initial fo
Month Feb Mar Apr May Jun Jul adjustment is 0. The smo
Income 70 68.5 64.8 71.7 71.3 72.8
b. Assume the smoothin
c. Using MSE, determine

Month Income a. Forecast a. Error2

Feb 70.00 65.00 25.00
Mar 68.50 65.90 6.76
Apr 64.80 66.77 3.87
May 71.70 67.02 21.89
Jun 71.30 68.31 8.91
Jul 72.80 69.68 9.76

MSE = 12.70
. Use Trend-adjusted exponential smoothing to forecast the firm's August income.
ssume that the initial forecast average for February is $65,000 and the initial trend
djustment is 0. The smoothing constants selected are α=.1 and β=.8

. Assume the smoothing constants selected are α=.1 and β=.3

. Using MSE, determine which smoothing constants provide a better forecast.
MSE = Σ(forecast errors)2 / n

a. Error2 b. Forecast a. Error2

=(70 - 65)^2 65.00 25.00 =(70 - 65)^2
=(68.5 - 65.9)^2 65.65 8.12 =(68.5 - 65.65)^2
=(64.8 - 66.77)^2 66.17 1.88 =(64.8 - 66.17)^2
=(71.7 - 67.02)^2 66.23 29.94 =(71.7 - 66.23)^2
=(71.3 - 68.31)^2 67.13 17.36 =(71.3 - 67.13)^2
=(72.8 - 69.68)^2 68.03 22.72 =(72.8 - 68.03)^2

17.50