CHAPTER 5

Forecasting
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, α, is high, more weight is given to recent data. When α 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 aver-
age of the absolute deviations. MAD is important because it can be used to help increase fore-
casting 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 in-
dex would reflect that.
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 (10 + 12 + 13 + 16)/4 = 51/4 = 12.75
June 23 (12 + 13 + 16 + 19)/4 = 60/4 = 15
July 26 (13 + 16 + 19 + 23)/4 = 70/4 = 17.75
Aug. 30 (16 + 19 + 23 + 26)/4 = 84/4 = 21
Sept. 28 (19 + 23 + 26 + 30)/4 = 98/4 = 24.5
Oct. 18 (23 + 26 + 30 + 28)/4 = 107/4 =
26.75
Nov. 16 (26 + 30 + 28 + 18)/4 = 102/4 = 25.5
Dec. 14 (30 + 28 + 18 + 16)/4 = 92/4 = 23

The MAD = 7.78

See solution to 5-13 for calculations.
5-14.
Three- Four-
Three- Month Four- Month
Actual Month Absolute Month Absolute
Month 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 7 15 8
July 26 19.33 6.67 17.75 8.25
Aug. 30 22.67 7.33 21 9
Sept. 28 26.33 1.67 24.5 3.5
Oct. 18 28 10 26.75 8.75
Nov. 16 25.33 9.33 25.5 9.5
Dec. 14 20.67 6.67 23 9
58.33 62.25

58.33
Three-month MAD = = 6.48
9
62.25
Four-month MAD = = 7.78
8
The 3-month moving average appears to be more accurate. However, when weighted moving av-
erages were used, the MAD was 5.444.
5-15.
Year Demand 3-Year Moving 3-Year Wt. Moving 3-Year Abs. 3-Year Wt. Abs.
Ave. Ave. Deviation Deviation
1 4
2 6
3 4
4 5 (4 + 6 + 4)/3 [(2 × 4) + 6 + 4]/4 0.34 0.55
= 4 23 = 4 12

5 10 (6 + 4 + 5)/3 [(2 × 5) + 4 + 6]/4 5 5

=5 =5

6 8 (4 + 5 + 10)/3 [(2 × 10) + 5 +4]/4 1.67 0.75

= 6 13 = 7 14

7 7 (5 + 10 + 8)/3 [(2 × 8) + 10 +5]/4 0.67 0.75

= 7 23 = 7 34

8 9 (10 + 8 + 7)/3 [(2 × 7) + 8 +10]/4 0.67 1

= 8 13 =8

9 12 (8 + 7 + 9)/3 [(2 × 9) + 7 + 8]/4 4 3.75

=8 = 8 14

10 14 (7 + 9 + 12)/3 [(2 × 12) + 9 +7]/4 4.67 4

= 9 13 = 10

11 15 (9 + 12 + 14)/3 [(2 × 14) + 12+9]/4 3.34 2.75

=11 2 3 = 12 1 4
Total absolute deviations: 20.36 18.55

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.
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.
3-Yr MA 3-Yr Wt. Trend line
MA
Year Demand |deviation| |deviation| |deviation|
1 4 — — 0.73
2 6 — — 1.67
3 4 — — 1.38
4 5 0.34 0.55 1.44
5 10 5.00 5.00 2.51
6 8 1.67 0.75 0.55
7 7 0.67 0.75 2.60
8 9 0.67 1.00 1.65
9 12 4.00 3.75 0.29
10 14 4.67 4.00 1.24
11 15 3.34 2.75 1.18
Total absolute deviations = 20.36 18.55 15.24

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. α = 0.3. New forecast for year 2 is last period’s forecast + α(last period’s actual demand −
last period’s 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
The calculations are:
Year Demand New Forecast
2 6,000 4,700 = 5,000 + (0.3)(4,000 − 5,000)
3 4,000 5,090 = 4,700 + (0.3)(6,000 − 4,700)
4 5,000 4,763 = 5,090 + (0.3)(4,000 − 5,090)
5 10,000 4,834 = 4,763 + (0.3)(5,000 − 4,763)
6 8,000 6,384 = 4,834 + (0.3)(10,000 − 4,834)
7 7,000 6,869 = 6,384 + (0.3)(8,000 − 6,384)
8 9,000 6,908 = 6,869 + (0.3)(7,000 − 6,869)
9 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
Moving Absolute Absolute
Year Demand Average Deviation Exp. Sm. Deviation
1 4,000 5,000 1,000
2 6,000 4,700 1,300
3 4,000 5,090 1,090
4 5,000 4,500 500 4,763 237
5 10,000 5,000 5,000 4,834 5,166
6 8,000 7,250 750 6,384 1,616
7 7,000 7,750 750 6,869 131
8 9,000 8,000 1,000 6,908 2,092
9 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 26,808
Mean: 2,312.5 2,437
Thus, the 3-year weighted moving average model appears to be more accurate.
5-19. α = 0.30
Year 1 2 3 4 5 6
Forecast 410.0 422.0 443.9 466.1 495.2 521.8
5-20.
Year Sales Forecast Using α = 0.6 Forecast Using α = 0.9
1 450 410
2 495 410 + (0.6)(450 − 410) = 434 410 + (0.9)(450 − 410) = 446
3 518 434 + (0.6)(495 − 434) = 470.6 446 + (0.9)(495 − 446) = 490.1
4 563 470.6 + (0.6)(518 − 470.6) = 499.0 490.1 + (0.9)(518 − 490.1) = 515.21
5 584 499 + (0.6)(563 − 499) = 537.4 515.21 + (0.9)(563 − 515.21) = 558.2
6 ? 537.4 + (0.6)(584 − 537.4) = 565.4 558.2 + (0.9)(584 − 558.2) = 581.4

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

MADα=0.3 = 372.8/5 = 74.56

MADα=0.6 = 259/5 = 51.8
MADα=0.9 = 190.5/5 = 38.1
Because it has the lowest MAD, the smoothing constant α = 0.9 gives the most accurate forecast.
5-22.
Year Sales Three-Year Moving Average
1 450
2 495
3 518
4 563 (450 + 495 + 518)/3 = 487.67
5 584 (495 + 518 + 563)/3 = 525.3
6 ? (518 + 563 + 584)/3 = 555
5-23.
Time
Period Sales
Year X Y X2 XY
1 1 450 1 450
2 2 495 4 990
3 3 518 9 1554
4 4 563 16 2252
5 5 584 25 2920
2,610 55 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.
Three-Year Moving Time-Series
Year Actual Average Forecast Absolute Dev- Forecast Absolute
Sales iation Deviation
1 450 — — 454.8 4.8
2 495 — — 488.4 6.6
3 518 — — 522.0 4.0
4 563 487.7 75.3 555.6 7.4
5 584 525.3 58.7 589.2 5.2
6 ? 555.0 — 622.8 —
Total absolute deviation 134.0 28.0

MADα=0.3 = 74.56 (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.
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 Month Demand Average Weighted Average
4 Apr. 10 13.67 14.5
5 May 15 13.33 12.67
6 June 17 13.67 13.5
7 July 11 14 15.17
8 Aug. 14 14.33 13.67
9 Sept. 17 14 13.50
10 Oct. 12 14 15
11 Nov. 14 14.33 14
12 Dec. 16 14.33 13.83
13 Jan. 11 14 14.67
14 Feb. – 13.67 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. α = 0.20
Sum of
Absolute
Actual Forecast Forecast
Week Miles (Ft) Error RSFE Errors MAD Track Signal
1 17 17.00 — — — — —
2 21 17.00 +4.00 +4.00 4.00 4.00 1
3 19 17.80 +1.20 +5.20 5.20 2.60 2
4 23 18.04 +4.96 +10.16 10.16 3.39 3
5 18 19.03 −1.03 +9.13 11.19 2.80 3.3
6 16 18.83 −2.83 +6.30 14.02 2.80 2.25
7 20 18.26 +1.74 +8.04 15.76 2.63 3.05
8 18 18.61 −0.61 +7.43 16.37 2.34 3.17
9 22 18.49 +3.51 +10.94 19.88 2.49 4.21
10 20 19.19 +0.81 +11.75 20.69 2.30 5.11
11 15 19.35 −4.35 +7.40 25.04 2.50 2.96
12 22 18.48 +3.52 +10.92 28.56 2.60 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.
5-27. a., b. See the accompanying table for a comparison of the calculations for the exponential-
ly 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 Smoothed Smoothed

Week, Value, Value, Forecast Value, Forecast
t Yt Ft(α = 0.1) Error Ft(α = 0.6) Error
1 50 50 — —
2 35 50.00 -15.00 50.00 -15.00
3 25 48.50 -23.50 41.00 -16.00
4 40 46.15 -6.15 31.40 8.60
5 45 45.54 -0.54 36.56 8.44
6 35 45.48 -10.48 41.62 -6.62
7 20 44.43 -24.43 37.65 -17.65
8 30 41.99 -11.99 27.06 2.94
9 35 40.79 -5.79 28.82 6.18
10 20 40.21 -20.21 32.53 -12.53
11 15 38.19 -23.19 25.01 -10.01
12 40 35.87 4.13 19.00 21.00
13 55 36.28 18.72 31.60 23.40
14 35 38.16 -3.16 45.64 -10.64
15 25 37.84 -12.84 39.26 -14.26
16 55 36.56 18.44 30.70 24.30
17 55 38.40 16.60 45.28 9.72
18 40 40.06 -0.06 51.11 -11.11
19 35 40.05 -5.05 44.45 -9.45
20 60 39.55 20.45 38.78 21.22
21 75 41.59 33.41 51.51 23.49
22 50 44.93 5.07 65.60 -15.60
23 40 45.44 -5.44 56.24 -16.24
24 65 44.90 20.10 46.50 18.50
25 46.91 57.60
5-28. Using data from Problem 5-27, with α = 0.9
Actual Smoothed
Week, Value, Value, Forecast
t Yt Ft(α = 0.9) Error
1 50 50 —
2 35 50.00 -15.00
3 25 36.50 -11.50
4 40 26.15 13.85
5 45 38.62 6.39
6 35 44.36 -9.36
7 20 35.94 -15.94
8 30 21.59 8.41
9 35 29.16 5.84
10 20 34.42 -14.42
11 15 21.44 -6.44
12 40 15.64 24.36
13 55 37.56 17.44
14 35 53.26 -18.26
15 25 36.83 -11.83
16 55 26.18 28.82
17 55 52.12 2.88
18 40 54.71 -14.71
19 35 41.47 -6.47
20 60 35.65 24.35
21 75 57.56 17.44
22 50 73.26 -23.26
23 40 52.33 -12.33
24 65 41.23 23.77
25 62.62
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.
5-29. Exponential smoothing with α = 0.1
Month Income Forecast Error
Feb. 70.0 65.0 —
March 68.5 65.0 + 0.1(70 − 65) = 65.5 3.0
April 64.8 65.5 + 0.1(68.5 − 65.5) = 65.8 −1.0
May 71.7 65.8 + 0.1(64.8 − 65.8) = 65.7 6.0
June 71.3 65.7 + 0.1(71.7 − 65.7) = 66.3 5.0
July 72.8 66.3 + 0.1(71.3 − 66.3) = 66.8 6.0
Aug. 66.8 + 0.1(72.8 − 66.8) = 67.4
MAD = 4.20

Note that in this problem, the initial forecast (for the first period) was not used in computing the
MAD. Either approach is considered valid.

5-30. Exponential smoothing with α = 0.3

Month Income Forecast Error
Feb. 70.0 65.0 —
March 68.5 66.5 2.0
April 64.8 67.1 −2.3
May 71.7 66.4 5.3
June 71.3 68.0 3.3
July 72.8 69.0 3.8
Aug. 70.1
MAD = 3.34

Based on MAD, α = 0.3 produces a better forecast than α = 0.1 (of Problem 5-29).
Note that in this problem, the initial forecast (for the first period) was not used in computing the
MAD. Either approach is considered valid.
5-31. Using QM for Windows, we select Forecasting - Time Series and multiplicative decompo-
sition. 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
 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
Q2 = 1 if quarter 2, 0 otherwise
Q3 = 1 if quarter 3, 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.
5-34. For a smoothing constant of 0.2, the forecast for year 11 is 6.489.
Year Rate Forecast |Error|
1 7.2 7.2 0
2 7 7.2 0.2
3 6.2 7.16 0.96
4 5.5 6.968 1.468
5 5.3 6.674 1.374
6 5.5 6.400 0.900
7 6.7 6.220 0.480
8 7.4 6.316 1.084
9 6.8 6.533 0.267
10 6.1 6.586 0.486
11 6.489
MAD = 0.722
For a smoothing constant of 0.4, the forecast for year 11 is 6.458.
Year Rate Forecast |Error|
1 7.2 7.2 0
2 7 7.2 0.2
3 6.2 7.12 0.92
4 5.5 6.752 1.252
5 5.3 6.251 0.951
6 5.5 5.871 0.371
7 6.7 5.722 0.978
8 7.4 6.113 1.287
9 6.8 6.628 0.172
10 6.1 6.697 0.597
11 6.458
MAD = 0.673
For a smoothing constant of 0.6, the forecast for year 11 is 6.401.
Year Rate Forecast |Error|
1 7.2 7.2 0
2 7 7.2 0.2
3 6.2 7.08 0.88
4 5.5 6.552 1.052
5 5.3 5.921 0.621
6 5.5 5.548 0.048
7 6.7 5.519 1.181
8 7.4 6.228 1.172
9 6.8 6.931 0.131
10 6.1 6.852 0.752
11 6.401
MAD = 0.604
For a smoothing constant of 0.8, the forecast for year 11 is 6.256.
Year Rate Forecast |Error|
1 7.2 7.2 0
2 7 7.2 0.2
3 6.2 7.04 0.84
4 5.5 6.368 0.868
5 5.3 5.674 0.374
6 5.5 5.375 0.125
7 6.7 5.475 1.225
8 7.4 6.455 0.945
9 6.8 7.211 0.411
10 6.1 6.882 0.782
11 6.256
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) =

( year 1 demand ) + ( year 2 demand )
2

Overall average demand =

( sum of all values )
8
( average for season )
Season index =
overall average demand
new annual demand
Year 3 demand =
4
1, 200
= × season index
4
Solution Table for Problem 5-36
Average
Year 1 Year 2 (Average Year 1- Season Season Year 3
Season Demand Demand Year 2 Demand) Demand Index Demand
Fall 200 250 225.0 250 0.90 270
Winter 350 300 325.0 250 1.30 390
Spring 150 165 157.5 250 0.63 189
Summer 300 285 292.5 250 1.17 351

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 α = 0.4, forecast for 2011 = 10338 and MAD = 836 (including first error). With α =
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.
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 α = 0.6, forecast for 2003 = 86.2 and MAD = 3.42. With α = 0.2, forecast for 2003 =
63.87 and MAD = 7.23. The model with α = 0.6 is better since it has a lower MAD.
5-48. With α = 0.6, forecast for 2003 = 4.86 and MAD = 0.23. With α = 0.2, forecast for 2003 =
4.52 and MAD = 0.48. The model with α = 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
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 de-
veloped 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
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% in-
crease) 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 evi-
denced 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.
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 Feb-
ruary 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 Unadjusted forecast Seasonal index Adjusted forecast

January 325.852 1.447 471.5
February 326.711 1.393 455.1
March 327.57 1.379 451.7
April 328.429 1.074 352.7
May 329.288 1.039 342.1
June 330.147 0.797 263.1
July 331.006 0.813 269.1
August 331.865 0.720 238.9
September 332.724 0.667 221.9
October 333.583 0.747 249.2
November 334.442 0.891 298.0
December 335.301 1.033 346.4

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.
 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 av-
erage 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.