Vous êtes sur la page 1sur 27

4

C H A P T E R

Forecasting

DISCUSSION QUESTIONS
1.Qualitative models incorporate subjective factors into the
forecasting model. Qualitative models are useful when subjective
factors are important. When quantitative data are difficult to
obtain, qualitative models may be appropriate.
2.Approaches are qualitative and quantitative. Qualitative is
relatively subjective; quantitative uses numeric models.
3.Short-range (under 3 months), medium-range (3 months to
3 years), and long-range (over 3 years).
4.The steps that should be used to develop a forecasting
system are:
(a)Determine the purpose and use of the forecast
(b)Select the item or quantities that are to be forecasted
(c)Determine the time horizon of the forecast
(d)Select the type of forecasting model to be used
(e)Gather the necessary data
(f) Validate the forecasting model
(g)Make the forecast
(h)Implement and evaluate the results
5.Any three of: sales planning, production planning and
budgeting, cash budgeting, analyzing various operating plans.
6.There is no mechanism for growth in these models; they are
built exclusively from historical demand values. Such methods
will always lag trends.
7.Exponential smoothing is a weighted moving average where
all previous values are weighted with a set of weights that decline
exponentially.
8.MAD, MSE, and MAPE are common measures of forecast
accuracy. To find the more accurate forecasting model, forecast
with each tool for several periods where the demand outcome is
known, and calculate MSE, MAPE, or MAD for each. The
smaller error indicates the better forecast.
9.The Delphi technique involves:
(a)Assembling a group of experts in such a manner as to
preclude direct communication between identifiable
members of the group
(b)Assembling the responses of each expert to the questions
or problems of interest
(c)Summarizing these responses
(d)Providing each expert with the summary of all responses

(e)Asking each expert to study the summary of the responses


and respond again to the questions or problems of interest.
(f) Repeating steps (b) through (e) several times as necessary
to obtain convergence in responses. If convergence has
not been obtained by the end of the fourth cycle, the
responses at that time should probably be accepted and
the process terminatedlittle additional convergence is
likely if the process is continued.
10.A time series model predicts on the basis of the assumption
that the future is a function of the past, whereas an associative
model incorporates into the model the variables of factors that
might influence the quantity being forecast.
11.A time series is a sequence of evenly spaced data points with the
four components of trend, seasonality, cyclical, and random variation.
12.When the smoothing constant, , is large (close to 1.0),
more weight is given to recent data; when is low (close to 0.0),
more weight is given to past data.
13.Seasonal patterns are of fixed duration and repeat regularly.
Cycles vary in length and regularity. Seasonal indices allow
generic forecasts to be made specific to the month, week, etc.,
of the application.
14.Exponential smoothing weighs all previous values with a set
of weights that decline exponentially. It can place a full weight on
the most recent period (with an alpha of 1.0). This, in effect, is the
nave approach, which places all its emphasis on last periods
actual demand.
15.Adaptive forecasting refers to computer monitoring of
tracking signals and self-adjustment if a signal passes its present
limit.
16.Tracking signals alert the user of a forecasting tool to
periods in which the forecast was in significant error.
17.The correlation coefficient measures the degree to which the
independent and dependent variables move together. A negative
value would mean that as X increases, Y tends to fall. The
variables move together, but move in opposite directions.
18.Independent variable (x) is said to explain variations in the
dependent variable (y).
19.Nearly every industry has seasonality. The seasonality must
be filtered out for good medium-range planning (of production
and inventory) and performance evaluation.
20.There are many examples. Demand for raw materials and
component parts such as steel or tires is a function of demand for
goods such as automobiles.

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

27

28

CHAPTER 4F O R E C A S T I N G

21.Obviously, as we go farther into the future, it becomes more


difficult to make forecasts, and we must diminish our reliance on
the forecasts.

*Active Models 4.1, 4.2, 4.3, and 4.4 appear on our Web site,
www.pearsonhighered.com/heizer.

ETHICAL DILEMMA
This exercise, derived from an actual situation, deals as much with
ethics as with forecasting. Here are a few points to consider:

No one likes a system they dont understand, and most


college presidents would feel uncomfortable with this one.
It does offer the advantage of depoliticizing the funds allocation if used wisely and fairly. But to do so means all
parties must have input to the process (such as smoothing
constants) and all data need to be open to everyone.
The smoothing constants could be selected by an agreedupon criteria (such as lowest MAD) or could be based on
input from experts on the board as well as the college.
Abuse of the system is tied to assigning alphas based on
what results they yield, rather than what alphas make the
most sense.
Regression is open to abuse as well. Models can use many
years of data yielding one result or few years yielding a
totally different forecast. Selection of associative variables
can have a major impact on results as well.

Active Model Exercises*


ACTIVE MODEL 4.1: Moving Averages
1.What does the graph look like when n = 1?
The forecast graph mirrors the data graph but one period
later.
2.What happens to the graph as the number of periods in the
moving average increases?
The forecast graph becomes shorter and smoother.
3.What value for n minimizes the MAD for this data?
n = 1 (a nave forecast)

ACTIVE MODEL 4.2: Exponential Smoothing


1.What happens to the graph when alpha equals zero?
The graph is a straight line. The forecast is the same in
each period.
2.What happens to the graph when alpha equals one?
The forecast follows the same pattern as the demand
(except for the first forecast) but is offset by one period. This
is a nave forecast.
3.Generalize what happens to a forecast as alpha increases.
As alpha increases the forecast is more sensitive to
changes in demand.
4.2(a)No, the data appear to have no consistent pattern
(see part d for graph).

(b)
(c)

Year

10

11

Forecast

Demand
3-year moving
3-year weighted

9.0
7.0
6.4

13.0
7.7
7.8

8.0
9.0
11.0

12.0
10.0
9.6

13.0
11.0
10.9

9.0
11.0
12.2

11.0
11.3
10.5

7.0
11.0
10.6

9.0
8.4

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

CHAPTER 4F O R E C A S T I N G

4.At what level of alpha is the mean absolute deviation (MAD)


minimized?
alpha = .16

29

(d)The three-year moving average appears to give better


results.

ACTIVE MODEL 4.3: Exponential Smoothing with


Trend Adjustment
1.Scroll through different values for alpha and beta. Which
smoothing constant appears to have the greater effect on the graph?
alpha
2.With beta set to zero, find the best alpha and observe the
MAD. Now find the best beta. Observe the MAD. Does the
addition of a trend improve the forecast?
4.3

Year

10

11

Forecast

Demand
Nave
Exp. Smoothing

9.0
7.0
6.4

5.0
9.0
7.4

9.0
5.0
6.5

13.0
9.0
7.5

8.0
13.0
9.7

12.0
8.0
9.0

13.0
12.0
10.2

9.0
13.0
11.3

11.0
9.0
10.4

7.0
11.0
10.6

7.0
9.2

alpha = .11, MAD = 2.59; beta above .6 changes the MAD


(by a little) to 2.54.

ACTIVE MODEL 4.4: Trend Projections


1.What is the annual trend in the data?
10.54
2.Use the scrollbars for the slope and intercept to determine the
values that minimize the MAD. Are these the same values that
regression yields?
No, they are not the same values. For example, an
intercept of 57.81 with a slope of 9.44 yields a MAD of 7.17.

Nave tracks the ups and downs best but lags the data by one
period. Exponential smoothing is probably better because it
smoothes the data and does not have as much variation.

END-OF-CHAPTER PROBLEMS

TEACHING NOTE: Notice


smoothing forecasts the nave.

374 + 368 + 381


4.1 (a)
= 374.33 pints
3

well

exponential

4.4 (a) FJuly = FJune + 0.2(Forecasting error)


= 42 + 0.2(40 42) = 41.6

(b)
Week of
August 31
September 7
September 14
September 21
September 28
October 5

how

Weighted
Moving Average

Pints Used
360
389
410
381
368
374
Forecast 372.9

(c)
The forecast is 374.26.

(b) FAugust = FJuly + 0.2(Forecasting error)


= 41.6 + 0.2(45 41.6) = 42.3

381 .1 = 38.1
368 .3 = 110.4
374 .6 = 224.4
372.9

Week of

Pints

Forecast

August 31
September 7
September 14
September 21
September 28
October 5

360
389
410
381
368
374

360
360
365.8
374.64
375.912
374.3296

Forecasting
Error
0
29
44.2
6.36
7.912
.3296

Error
.20
0
5.8
8.84
1.272
1.5824
.06592

Forecast
360
365.8
374.64
375.912
374.3296
374.2636

(c)The banking industry has a great deal of seasonality in


its processing requirements
4.5 (a)

3,700 + 3,800
=3,750 miles
2

(b)
Year
1
2

Mileage

Two-Year
Moving Average

3,000
4,000

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

Error

|Error|

30

3
4
5

CHAPTER 4F O R E C A S T I N G

3,400
3,800
3,700

3,500
3,700
3,600
Totals

MAD =

300
= 100.
3

100
100
100
100

100
100
100
300

4.5(c)Weighted 2 year M.A. with .6 weight for most recent year.


Year
1
2
3
4
5

Mileage

Forecast

Error

3,000
4,000
3,400
3,800
3,700

3,600
3,640
3,640

200
160
60

|Error|

200
160
60
420

Forecast for year 6 is 3,740 miles.

420
MAD = 140 =

3
4.5(d)
Year

Mileage

Forecast

Forecast
Error

Error
= .50

New
Forecast

3,000
4,000
3,400
3,800
3,700

3,000
3,000
3,500
3,450
3,625
Total

0
1,000
100
350
75
1,325

0
500
50
175
38

3,000
3,500
3,450
3,625
3,663

1
2
3
4
5

The forecast is 3,663 miles.

4.6
January
February
March
April
May
June
July
August
September
October
November
December
Sum
Average

Y Sales

X Period

20
21
15
14
13
16
17
18
20
20
21
23

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

218
18.2

(a)

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

78
6.5

X2
1
4
9
16
25
36
49
64
81
100
121
144

XY
20
42
45
56
65
96
119
144
180
200
231
276

650

1,474

CHAPTER 4F O R E C A S T I N G

(b)[i]Nave

The coming January = December = 23

[ii]3-month moving (20 + 21 + 23)/3 = 21.33


[iii]6-month weighted [(0.1 17) + (.1 18)
+ (0.1 20) + (0.2 20)
+ (0.2 21) + (0.3 23)]/1.0
= 20.6
[iv]Exponential smoothing with alpha = 0.3
FOct = 18 + 0.3(20 18) = 18.6
FNov = 18.6 + 0.3(20 18.6) = 19.02
FDec = 19.02 + 0.3(21 19.02) = 19.6
FJan = 19.6 + 0.3(23 19.6) = 20.62 21
[v]Trend x = 78, x = 6.5, y = 218, y = 18.17
xy nx y
x 2 nx 2
1474 (12)(6.5)(18.2) 54.4
b=
=
= 0.38
650 12(6.5)2
143
a = y bx
a = 18.2 0.38(6.5) = 15.73
b=

Forecast = 15.73+.38(13) = 20.67, where next January


is the 13th month.
(c)Only trend provides an equation that can extend beyond
one month
4.7Present = Period (week) 6.
1
1
1
1
a) So: F7 = 3 A6 + 4 A5 + 4 A4 + 6 A3 1.0




1
1
1
1
= (52) + (63) + (48) + (70) = 56.76 patients,
3
4
4
6
or 57 patients
1 1 1 1
where 1.0 = weights , , ,
3 4 4 6
b) If the weights are 20, 15, 15, and 10, there will be no change in
the forecast because these are the same relative weights as in
part (a), i.e., 20/60, 15/60, 15/60, and 10/60.
c) If the weights are 0.4, 0.3, 0.2, and 0.1, then the forecast
becomes 56.3, or 56 patients.

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

31

32

CHAPTER 4F O R E C A S T I N G

Table for Problem 4.9 (a, b, c)


Forecast

Month
January
February
March
April
May
June
July
August
September
October
November
December

Price per
Chip
$1.80
1.67
1.70
1.85
1.90
1.87
1.80
1.83
1.70
1.65
1.70
1.75

(96 + 88 + 90)
= 91.3
3
(88 + 90)
(b)
= 89
2
(c)

4.8 (a)

93.5
93.5
94.0
95.5
92.0

1.735
1.685
1.775
1.875
1.885
1.835
1.815
1.765
1.675
1.675

Three-Month
Moving
Average

Two-Month
Moving
Average

1.723
1.740
1.817
1.873
1.857
1.833
1.777
1.727
1.683
Totals

.035
.165
.125
.005
.085
.005
.115
.115
.025
.075
.750

|Error|
Three-Month
Moving
Average

.127
.160
.053
.073
.027
.133
.127
.027
.067
.793

Therefore, the two-month moving average seems to have


performed better.

Temperature 2 day M.A. |Error| (Error)2


93
94
93
95
96
88
90

Two-Month
Moving
Average

0.5 0.25
1.5
2.25
2.0 4.00
7.5
56.25
2.0 4.00
13.5
66.75

Absolute % Error

100(.5/93)
100(1.5/95)
100(2/96)
100(7.5/88)
100(2/90)

= 0.54%
= 1.58%
= 2.08%
= 8.52%
= 2.22%
14.94%

MAD = 13.5/5 = 2.7

(d)MSE = 66.75/5 = 13.35


(e)MAPE = 14.94%/5 = 2.99%
4.9(a, b) The computations for both the two- and three-month
averages appear in the table; the results appear in the
figure below.

(c)MAD (two-month moving average) = .750/10 = .075


MAD (three-month moving average) = .793/9 = .088
Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

CHAPTER 4F O R E C A S T I N G
4.9

(d)Table for Problem 4.9(d):


= .1
Month

Price per Chip

Forecast

January
February
March
April
May
June
July
August
September
October
November
December

$1.80
1.67
1.70
1.85
1.90
1.87
1.80
1.83
1.70
1.65
1.70
1.75
Totals
MAD (total/12)

$1.80
1.80
1.79
1.78
1.79
1.80
1.80
1.80
1.81
1.80
1.78
1.77

= .3
|Error|

Forecast

$.00
.13
.09
.07
.11
.07
.00
.03
.11
.15
.08
.02
$.86
$.072

$1.80
1.80
1.76
1.74
1.77
1.81
1.83
1.82
1.82
1.79
1.75
1.73

= .5
|Error|

Forecast

$.00
.13
.06
.11
.13
.06
.03
.01
.12
.14
.05
.02
$.86
$.072

$1.80
1.80
1.74
1.72
1.78
1.84
1.86
1.83
1.83
1.76
1.71
1.70

|Error|
$.00
.13
.04
.13
.12
.03
.06
.00
.13
.11
.01
.05
$.81
$.0675

= .5 is preferable, using MAD, to = .1 or = .3. One could


also justify excluding the January error and then dividing by
n = 11 to compute the MAD. These numbers would be $.078
(for = .1), $.078 (for = .3), and $.074 (for = .5).
4.10

Year
Demand
(a) 3-year moving
(b) 3-year weighted

10

11

Forecast

5.0
4.7
4.5

10.0
5.0
5.0

8.0
6.3
7.3

7.0
7.7
7.8

9.0
8.3
8.0

12.0
8.0
8.3

14.0
9.3
10.0

15.0
11.7
12.3

13.7
14.0

(c)The forecasts are about the same.


4.11 (a) Year
Demand
Exp. Smoothing

10

11

Forecast

4
5

6.0
4.7

4.0
5.1

5.0
4.8

10.0
4.8

8.0
6.4

7.0
6.9

9.0
6.9

12.0
7.5

14.0
8.9

15.0
10.4

11.8

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

33

34

CHAPTER 4F O R E C A S T I N G

(b) |Error| = |Actual Forecast|


Year
Exp. smoothing

1
1

2
1.3

3
1.1

4
0.2

5
5.2

6
1.6

7
0.1

8
2.1

9
4.5

10
5.1

11
4.6

MAD
2.4

These calculations were completed in Excel. Calculations are slightly different in Excel OM and POM for Windows due to
rounding differences.

4.12

MAD = 6.2
(c)Trend projection:

Actual
Demand

Forecast
Demand

Monday

88

88

Year

Tuesday

72

88

Wednesday

68

84

Thursday

48

80

Friday

1
2
3
4
5
6

Day

72

Answer

Ft = Ft1 + (At1 Ft1)

Demand

Trend Projection

45
50
52
56
58
?

42.6 + 3.2 1 = 45.8


42.6 + 3.2 2 = 49.0
42.6 + 3.2 3 = 52.2
42.6 + 3.2 4 = 55.4
42.6 + 3.2 5 = 58.6
42.6 + 3.2 6 = 61.8

Y = a + bX

F3 = 88 + .25(72 88) = 88 4 = 84

b=

F4 = 84 + .25(68 84) = 84 4 = 80
4.13(a)Exponential smoothing, = 0.6:
Year
1
2
3
4
5
6

Demand
45
50
52
56
58
?

41
41.0 + 0.6(4541) = 43.4
43.4 + 0.6(5043.4) = 47.4
47.4 + 0.6(5247.4) = 50.2
50.2 + 0.6(5650.2) = 53.7
53.7 + 0.6(5853.7) = 56.3

Absolute
Deviation
4.0
6.6
4.6
5.8
4.3

= 25.3
MAD = 5.06
Exponential smoothing, = 0.9:
Year
1
2
3
4
5
6

Exponential
Smoothing = 0.9

Demand
45
50
52
56
58
?

41
41.0 + 0.9(4541) = 44.6
44.6 + 0.9(5044.6 ) = 49.5
49.5 + 0.9(5249.5) = 51.8
51.8 + 0.9(5651.8) = 55.6
55.6 + 0.9(5855.6) = 57.8

Absolute
Deviation
4.0
5.4
2.5
4.2
2.4

= 18.5
MAD = 3.7
(b)3-year moving average:
Year
1
2
3
4
5
6

Demand
45
50
52
56
58
?

Three-Year
Moving Average

XY nXY
X 2 nX 2

a = Y bX

F5 = 80 + .25(48 80) = 80 8 = 72
Exponential
Smoothing = 0.6

0.8
1.0
0.2
0.6
0.6

= 3.2
MAD = 0.64

Let = .25. Let Monday forecast demand = 88


F2 = 88 + .25(88 88) = 88 + 0 = 88

Absolute
Deviation

Absolute
Deviation

XY

X2

1
2
3
4
5

45
50
52
56
58

45
100
156
224
290

1
4
9
16
25

Then: X = 15, Y = 261, XY = 815, X2 = 55, X = 3, Y = 52.2


Therefore:
815 5 3 52.2
b=
= 3.2
55 5 3 3
a = 52.20 3.20 3 = 42.6
Y6 = 42.6 + 3.2 6 = 61.8
(d) Comparing the results of the forecasting methodologies
for parts (a), (b), and (c).
Forecast Methodology

MAD

Exponential smoothing, = 0.6


Exponential smoothing, = 0.9
3-year moving average
Trend projection

5.06
3.7
6.2
0.64

Based on a mean absolute deviation criterion, the trend projection


is to be preferred over the exponential smoothing with = 0.6,
exponential smoothing with = 0.9, or the 3-year moving average
forecast methodologies.
4.14
Method 1:

(45 + 50 + 52)/3 = 49
(50 + 52 + 56)/3 = 52.7
(52 + 56 + 58)/3 = 55.3

7
5.3

MAD: (0.20 + 0.05 + 0.05 + 0.20)/4 = .125 better


MSE : (0.04 + 0.0025 + 0.0025 + 0.04)/4 = .021

Method 2:

MAD: (0.1 + 0.20 + 0.10 + 0.11) / 4 = .1275


MSE : (0.01 + 0.04 + 0.01 + 0.0121) / 4 = .018 better

= 12.3
Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

CHAPTER 4F O R E C A S T I N G

35

4.15
Year

Sales

2005
2006
2007
2008
2009
2010

450
495
518
563
584

Forecast Three-Year
Moving Average

Absolute
Deviation

(450 + 495 + 518)/3 = 487.7


(495 + 518 + 563)/3 = 525.3
(518 + 563 + 584)/3 = 555.0

75.3
58.7

Year

Sales

2005
2006
2007
2008
2009
2010

450
495
518
563
584

Forecast Exponential
Smoothing = 0.9
410.0
410 + 0.9(450 410) = 446.0
446 + 0.9(495 446) = 490.1
490.1 + 0.9(518 490.1) = 515.2
515.2 + 0.9(563 515.2) = 558.2
558.2 + 0.9(584 558.2) = 581.4

= 134
MAD = 67
4.16
Year

Time Period X

Sales Y

X2

2005
2006
2007
2008
2009

1
2
3
4
5

450
495
518
563
584

1
4
9
16
25

450
990
1554
2252
2920

= 2610

= 55

= 8166

XY

X = 3, Y = 522
Y = a + bX
b=

XY nXY
X 2 nX 2

8166 (5)(3)(522) 336


=
= 33.6
55 (5)(9)
10

a = Y bX = 522 (33.6)(3) = 421.2


y = 421.2 + 33.6 x
y = 421.2 + 33.6 6 = 622.8
Sales

Forecast Trend

Absolute Deviation

2005
2006
2007
2008
2009
2010

450
495
518
563
584

454.8
488.4
522.0
555.6
589.2
622.8

4.8
6.6
4.0
7.4
5.2

2005
2006
2007
2008
2009
2010

450
495
518
563
584

MAD= 0.3 = 74.6


MAD= 0.6 = 51.8
MAD= 0.9 = 38.1
Because it gives the lowest MAD, the smoothing constant of
= 0.9 gives the most accurate forecast.
4.18We need to find the smoothing constant . We know in
general that Ft = Ft1 + (At1 Ft1); t = 2, 3, 4. Choose
either
t = 3 or t = 4 (t = 2 wont let us find because F2 = 50 = 50 +
(50 50) holds for any ). Lets pick t = 3. Then F3 = 48 = 50 +
(42 50)
48 = 50 + 42 50

or

2 = 8

So,

.25 =
F5 = 50 + 46 50 = 50 4

For

= .25, F5 = 50 4(.25) = 49

The forecast for time period 5 = 49 units.


4.19Trend adjusted exponential smoothing: = 0.1, = 0.2
Month

4.17
Sales

(Refer to Solved Problem 4.1)


For = 0.3, absolute deviations for 20052009 are 40.0, 73.0,
74.1, 96.9, 88.8, respectively. So the MAD = 372.8/5 = 74.6.

Now we can find F5 : F5 = 50 + (46 50)

= 28
MAD = 5.6

Year

40.0
49.0
27.9
47.8
25.8

= 190.5
MAD = 38.1

or

Year

Absolute
Deviation

Forecast Exponential
Smoothing = 0.6
410.0
410 + 0.6(450 410) = 434.0
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) = 565.6

Absolute
Deviation
40.0
61.0
47.4
64.0
46.6

= 259
MAD = 51.8

February
March
April
May
June
July
August

Income
70.0
68.5
64.8
71.7
71.3
72.8

Unadjusted
Forecast
65.0
65.5
65.9
65.92
66.62
67.31
68.16

Trend
0.0
0.1
0.16
0.13
0.25
0.33

Adjusted
Forecast |Error| Error 2
65
65.6
66.05
66.06
66.87
67.64
68.60

5.0
2.9
1.2
5.6
4.4
5.2
24.3

25.0
8.4
1.6
31.9
19.7
26.6
113.2

MAD = 24.3/6 = 4.05, MSE = 113.2/6 = 18.87. Note that all


numbers are rounded.
Note: To use POM for Windows to solve this problem, a period 0,
which contains the initial forecast and initial trend, must be added.

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

36

CHAPTER 4F O R E C A S T I N G

4.20Trend adjusted exponential smoothing: = 0.1, = 0.8


4.23Students must determine the nave forecast for the four

4.21 F5 = A4 + ( 1 ) ( F4 + T4 ) = ( 0.2Unadjusted
) ( 19) + ( 0.8) ( 20.14)

Month

Demand (y)

Forecast

= 3.8 + 16.11 = 19.91

Trend

February
70.0
65.0
0
March T5 = F5 F4 + 1 68.5
65.5 17.82 0.4
T4 = 0.4 19.91
April
64.8
66.16
0.61
+ 0.666.57
2.32 = 0.40.45
2.09
May
71.7
June
71.3
+ 1.3967.49
= 0.84 + 1.390.82
= 2.23
July
72.8
68.61
1.06
+ 2.23 = 22.14
Totals FIT5 = F5 + T5 = 19.91
419.1
Average
69.85
August Forecast

) (

( )(
( )(

Adjusted
months. The
nave |Error|
forecast for
March is the February actual of 83,
Forecast
Error
Error2

etc.

65.0
65.9
66.77
67.02
68.31
69.68

5.00
2.60
1.97
4.68
2.99
3.12
16.42
2.74
(Bias)

5.0
2.6
1.97
4.68
2.99
3.12
20.36
3.39
(MAD)

25.00
6.76
3.87
21.89
8.91
9.76
76.19
12.70
(MSE)

71.30
F6 = A5 + ( 1 ) ( F5 + T5 ) = ( 0.2 ) ( 24 ) + ( 0.8) ( 22.14 )
Based upon the MSE criterion, =the
smoothing with = 0.1, = 0.8 is to be preferred
4.8 exponential
+ 17.71 = 22.51
over the exponential smoothing with = 0.1, = 0.2. Its MSE of 12.70 is lower. Its MAD of 3.39 is
also lower
T6 = than
( F6that
F5in) +Problem
= 0.4 ( 22.51 19.91) + 0.6 ( 2.23)
( 1 ) T54.19.

= 0.4 ( 2.6 ) + 1.34

= 1.04 + 1.34 = 2.38

March
April
May
June

FIT6 = F6 + T6 = 22.51 + 2.38 = 24.89


4.22 F7 = A6 + (1 )( F6 + T6 ) = (0.2)(21) + (0.8)(24.89)
= 4.2 + 19.91 = 24.11
T7 = ( F7 F6 ) + (1 )T6 = (0.4)(24.11 22.51)

+ (0.8)(26.18) = 27.14
T8 = ( F8 F7 ) + ( 1 ) T7 = 0.4 ( 27.14 24.11)

+ 0.6 ( 2.07 ) = 2.45

FIT8 = F8 + T8 = 27.14 + 2.45 = 29.59


F9 = A8 + ( 1 ) ( F8 + T8 ) = ( 0.2 ) ( 28)

+ ( 0.8) ( 29.59 ) = 29.28

T9 = ( F9 F8 ) + ( 1 ) T8 = ( 0.4 ) ( 29.28 27.14 )

+ ( 0.6 ) ( 2.45) = 2.32

FIT9 = F9 + T9 = 29.28 + 2.32 = 31.60

101
96
89
108

|Error| |% Error|

120
114
110
108

19
18
21
0
58

100 (19/101) = 18.81%


100 (18/96) = 18.75%
100 (21/89) = 23.60%
100 (0/108) = 0%
61.16%

58
= 14.5
4
61.16%
MAPE (for management) =
= 15.29%
4
MAD (for management) =

+ (0.6)(2.38) = 2.07
FIT7 = F7 + T7 = 24.11 + 2.07 = 26.18
F8 = A7 + (1 )( F7 + T7 ) = (0.2)(31)

Actual Forecast

(a)

(b)
March
April
May
June

Actual

Nave

101
96
89
108

83
101
96
89

|Error| |% Error|
18
5
7
19
49

100 (18/101) = 17.82%


100 (5/96) = 5.21%
100 (7/89) = 7.87%
100 (19/108) = 17.59%
48.49%

49
= 12.25
4
48.49%
MAPE (for nave) =
= 12.12%.
4
Nave outperforms management.
MAD (for nave) =

(c)MAD for the managers technique is 14.5, while MAD for the
nave forecast is only 12.25. MAPEs are 15.29% and 12.12%,
respectively. So the nave method is better.
4.24(a)Graph of demand
The observations obviously do not form a straight line but do tend
to cluster about a straight line over the range shown.
(b)Least-squares regression:
Y = a + bX
b=

XY nXY
X 2 nX 2

a = Y bX
Assume
Appearances X
3
4
7

Demand Y
3
6
7

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

X2

Y2

XY

9
16
49

9
36
49

9
24
49

CHAPTER 4F O R E C A S T I N G

6
8
5
9

5
10
7
?

36
64
25

25
100
49

30
80
35

X = 33, Y = 38, XY = 227, X2 = 199, X = 5.5, Y = 6.33.


Therefore:
227 6 5.5 6.333
= 1.0286
199 6 5.5 5.5
a = 6.333 1.0286 5.5 = .6762
Y = .676 + 1.03 x (rounded)
b=

b=

xy n x y

650 4(2.5)(55)

x nx
100
=
= 20
5
a = y bx
= 55 (20)(2.5)
=5
2

30 4(2.5)

37

650 550
30 25

The regression line is y = 5 + 20x. The forecast for May (x = 5) is


y = 5 + 20(5) = 105.

The following figure shows both the data and the resulting equation:

4.26
Average
Year1
Year2 Year1Year2
Season Demand Demand Demand
Fall
Winter
Spring
Summer

200
350
150
300

250
300
165
285

Average
Season Seasonal Year3
Demand Index Demand

225.0
325.0
157.5
292.5

250
250
250
250

0.90
1.30
0.63
1.17

270
390
189
351

Average Yr1 to Yr2 Yr1 Demand + Yr2 Demand

=
2
Demand for season
Sum of Ave Yr1 to Yr2 Demand
4
Average Yr1 to Yr2 Demand
Seasonal index =
Average Seasonal Demand
New Annual Demand
Yr3 =
Seasonal index
20,000 4
Average over all seasons:
= 1, 250
120016
=
Seasonal index
6,000
4 = 1,500
Average over spring:
4
1,500
Spring index:
= 1.2
1,250

Average seasonal demand =

(c) If there are nine performances by Stone Temple


Pilots, the estimated sales are:
Y9 = .676 + 1.03 9 = .676 + 9.27 = 9.93 drums
10 drums

(d) R = .82 is the correlation coefficient, and R 2 = .68


means 68% of the variation in sales can be explained by TV
appearances.

5,600
Answer:
(1.2) = 1,680 sailboats
4

4.25

Month

Number of
Accidents
(y)

January
February
March
April
Totals
Averages

30
40
60
90
220
y = 55

x
1
2
3
4
10
x = 2.5

xy

x2

30
80
180
360
650

1
4
9
16
30

4.27

2006
2007
2008
2009

Winter

Spring

Summer

1,400
1,200
1,000
900
4,500

1,500
1,400
1,600
1,500
6,000

1,000
2,100
2,000
1,900
7,000

Fall
600
750
650
500
2,500

4.28

Quarter

2007

2008

2009

Winter
Spring
Summer
Fall

73
104
168
74

65
82
124
52

89
146
205
98

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

Average
Average Quarterly Seasonal
Demand Demand
Index
75.67
110.67
165.67
74.67

106.67
106.67
106.67
106.67

0.709
1.037
1.553
0.700

4.32(a)

38

CHAPTER 4F O R E C A S T I N G

4.292011 is 25 years beyond 1986. Therefore, the 2011 quarter


numbers are 101 through 104.

(1)
Quarter

(2)
Quarter
Number

(3)
Forecast
(77 + .43Q)

(4)
Seasonal
Factor

(5)
Adjusted
Forecast
[(3) (4)]

Winter
Spring
Summer
Fall

101
102
103
104

120.43
120.86
121.29
121.72

.8
1.1
1.4
.7

96.344
132.946
169.806
85.204

Y = 36 + 4.3(70) = 337

(b)

Y = 36 + 4.3(80) = 380

(c)

Y = 36 + 4.3(90) = 423

4.31

xi

yi

i =1

330
270
380
300

5,280
3,240
6,840
4,200

256
144
324
196

60

1,280

19,560

920

(b)MSE = 160/5 = 32
(c)MAPE = 13.23%/5 = 2.65%
(1)

= Average number sold = 550

(2)

4.34Y = 7.5 + 3.5X1 + 4.5X2 + 2.5X3


(a)28
(b)43

Y=

16
12
18
14

(b)If the forecast is for 20 guests, the bar sales forecast


is 50 + 18(20) = $410. Each guest accounts for an
additional $18 in bar sales.

= Average price = 3.2583

x2

60
= 15
4
1,280
y=
= 320
4
xy nx y 19,560 4(15)(320) 360
b=
=
=
= 18
20
x 2 nx 2
920 4(15)2
a = y bx = 320 18(15) = 50
Y = 50 + 18 x

i =1

xy

x=

Let x1, x2 , K , x6 be the prices and y1, y2 , K , y6


be the number sold.
X=

So at x = 2.80, y = 1,454.6 277.6($2.80) = 677.32. Now round to


the nearest integer: Answer: 677 lattes.

4.30Given Y = 36 + 4.3X
(a)

4.33(a)See the table below.

xi yi = 9,783

(3)

i =1
6

b=

2
xi = 67.1925

Then y = a + bx, where y = number sold, x = price, and


6

b=

i =1
6

2
2
x i nx

2,880 2,700
55 45

55 5(3)
180
=
= 18
10
a = 180 3(18) = 180 54 = 126
y = 126 + 18 x

(4)

i=1

xi yi nx y

2,880 5(3)(180)

(9, 783) 6(3.25833)(550)

67.1925 6(3.25833)2

i =1

969.489
=
= 277.6
3.49222
a = y bx = 550 [(277.6)(3.25)] = 1, 454.6
Table for Problem 4.33
Year
(x)
1
2
3
4
5
Totals 15
x = 3

Transistors
(y)
140
160
190
200
210
900
y = 180

xy

x2

126 + 18x

Error

Error 2

|% Error|

140
320
570
800
1,050
2,800

1
4
9
16
25
55

144
162
180
198
216

4
2
10
2
6

16
4
100
4
36
160

100 (4/140) = 2.86%


100 (2/160) = 1.25%
100 (10/190) = 5.26%
100 (2/200) = 1.00%
100 (6/210) = 2.86%
13.23%

(c)58
4.35(a) Y = 13,473 + 37.65(1860) = 83,502
(b)The predicted selling price is $83,502, but this is the
average price for a house of this size. There are other
factors besides square footage that will impact the
selling price of a house. If such a house sold for
$95,000, then these other factors could be contributing
to the additional value.
Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

CHAPTER 4F O R E C A S T I N G

(c)Some other quantitative variables would be age of the


house, number of bedrooms, size of the lot, and size of
the garage, etc.
(d)Coefficient of determination = (0.63)2 = 0.397. This
means that only about 39.7% of the variability in the
sales price of a house is explained by this regression
model that only includes square footage as the
explanatory variable.
4.36(a)Given: Y = 90 + 48.5X1 + 0.4X2 where:
Y = expected travel cost
X1 = number of days on the road

1
2
3
4
5
6
7
8
9
10
55

36
33
40
41
40
55
60
54
58
61
478

b=

r = 0.68 (coefficient of correlation)


If:
Number of days on the road X1 = 5 and distance
traveled X2 = 300
then:
Y = 90 + 48.5 5 + 0.4 300 = 90 + 242.5 + 120 = 452.5
Therefore, the expected cost of the trip is $452.50.
(b)The reimbursement request is much higher than
predicted by the model. This request should probably
be questioned by the accountant.

b=

2900 10 5.5 47.8

385 10 5.5
a = 47.8 3.28 5.5 = 29.76
2

2900 2629 271


=
= 3.28
385 302.5 82.5

and Y = 29.76 + 3.28X. For:


X = 11: Y = 29.76 + 3.28 11 = 65.8
X = 12: Y = 29.76 + 3.28 12 = 69.1

3.costs of entertaining customers

Year 11 65.8 patients

4.other transportation costscab, limousine, special


tolls, or parking

Year 12 69.1 patients

In addition, the correlation coefficient of 0.68 is not exceptionally


high. It indicates that the model explains approximately 46% of
the overall variation in trip cost. This correlation coefficient
would suggest that the model is not a particularly good one.
4.37 (a, b)
20
21
28
37
25
29
36
22
25
28

20
20
20.5
24.25
30.63
27.81
28.41
32.20
27.11
26.05

Error
0.00
1.00
7.50
12.75
5.63
1.19
7.59
10.20
2.10
1.95

Running sum

|error|

0.00
1.00
7.50
12.75
5.63
1.19
7.59
10.20
2.10
1.95
MAD
5.00
Cumulative error = 14.05; MAD = 5Tracking = 14.05/5 = 2.82

0.00
1.00
8.50
21.25
15.63
16.82
24.41
14.21
12.10
14.05

4.38(a)least squares equation: Y = 0.158 + 0.1308X


(b)Y = 0.158 + 0.1308(22) = 2.719 million
(c)coefficient of correlation = r = 0.966
coefficient of determination = r2 = 0.934
4.39
Year X

X 2 nX 2

Therefore:

2.conference fees, if any

1
2
3
4
5
6
7
8
9
10

XY nXY

and X = 55, Y = 478, XY = 2900, X 2 = 385, Y 2 = 23892,


X = 5.5, Y = 47.8, Then:

1.the type of travel (air or car)

Forecast

Patients Y

X2

36
66
120
164
200
330
420
432
522
610
2,900

a = Y bX

(c)A number of other variables should be included, such as:

Demand

1,296
1,089
1,600
1,681
1,600
3,025
3,600
2,916
3,364
3,721
23,892

Given: Y = a + bX where:

X2 = distance traveled, in miles

Period

1
4
9
16
25
36
49
64
81
100
385

39

Y2

The model seems to fit the data pretty well. One should,
however, be more precise in judging the adequacy of the model.
Two possible approaches are computation of (a) the correlation
coefficient, or (b) the mean absolute deviation. The correlation
coefficient:
r=

n XY X Y
n X 2 ( X ) 2 n Y 2 ( Y ) 2

10 2900 55 478
10 385 552 10 23892 4782

29000 26290

3850 3025 238920 228484


2710
2710
=
=
= 0.924
2934.3
825 10436
r 2 = 0.853
The coefficient of determination of 0.853 is quite respectable
indicating our original judgment of a good fit was appropriate.

XY

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

40

Year
X

CHAPTER 4F O R E C A S T I N G

Patients
Y

Trend
Forecast

Deviation

Absolute
Deviation

36
33
40
41
40
55
60
54
58
61

29.8 + 3.28 1 = 33.1


29.8 + 3.28 2 = 36.3
29.8 + 3.28 3 = 39.6
29.8 + 3.28 4 = 42.9
29.8 + 3.28 5 = 46.2
29.8 + 3.28 6 = 49.4
29.8 + 3.28 7 = 52.7
29.8 + 3.28 8 = 56.1
29.8 + 3.28 9 = 59.3
29.8 + 3.28 10 = 62.6

2.9
3.3
0.4
1.9
6.2
5.6
7.3
2.1
1.3
1.6

2.9
3.3
0.4
1.9
6.2
5.6
7.3
2.1
1.3
1.6

1
2
3
4
5
6
7
8
9
10

Note that rounding differences occur when solving with Excel.


4.41note
(a)It
appears
from
graph that
points
do
Also
that
a crime
ratethe
of following
131.2 is outside
the the
range
of the
scatter
around a straight
line. equations, so caution is
data set used
to determine
the regression
advised.

= 32.6
MAD = 3.26
The MAD is 3.26this is approximately 7% of the average
number of patients and 10% of the minimum number of patients.
We also see absolute deviations, for years 5, 6, and 7 in the range
5.67.3. The comparison of the MAD with the average and
minimum number of patients and the comparatively large
deviations during the middle years indicate that the forecast model
is not exceptionally accurate. It is more useful for predicting
general trends than the actual number of patients to be seen in a
specific year.
4.40
Year
1
2
3
4
5
6
7
8
9
10
Column Totals

Crime
Rate X

Patients
Y

X2

Y2

XY

58.3
61.1
73.4
75.7
81.1
89.0
101.1
94.8
103.3
116.2
854.0

36
33
40
41
40
55
60
54
58
61
478

3,398.9
3,733.2
5,387.6
5,730.5
6,577.2
7,921.0
10,221.2
8,987.0
10,670.9
13,502.4
76,129.9

1,296
1,089
1,600
1,681
1,600
3,025
3,600
2,916
3,364
3,721
23,892

2,098.8
2,016.3
2,936.0
3,103.7
3,244.0
4,895.0
6,066.0
5,119.2
5,991.4
7,088.2
42,558.6

Given: Y = a + bX where
b=

XY nXY
X nX
2

(b)Developing the regression relationship, we have:


(Summer
months)
Year

42558.6 10 85.4 47.8

76129.9 10 85.42
1737.4
=
= 0.543
3197.3
a = 47.8 0.543 85.4 = 1.43
and
For:

42558.6 40821.2
76129.9 72931.6

Y = 1.43 + 0.543X
X = 131.2 : Y = 1.43 + 0.543(131.2) = 72.7
X = 90.6 : Y = 1.43 + 0.543(90.6) = 50.6

X2

Y2

XY

7
2
6
4
14
15
16
12
14
20
15
7

1.5
1.0
1.3
1.5
2.5
2.7
2.4
2.0
2.7
4.4
3.4
1.7

49
4
36
16
196
225
256
144
196
400
225
49

2.25
1.00
1.69
2.25
6.25
7.29
5.76
4.00
7.29
19.36
11.56
2.89

10.5
2.0
7.8
6.0
35.0
40.5
38.4
24.0
37.8
88.0
51.0
11.9

Given: Y = a + bX where:
b=

Crime rate = 131.2 72.7 patients


Crime rate = 90.6 50.6 patients

X 2 nX 2

and X = 132, Y = 27.1, XY = 352.9, X2 = 1796,


Y 2 = 71.59, X = 11, Y = 2.26.
Then:
352.9 298.3 54.6
=
=
= 0.159
1796 1452
344
1796 12 112
a = 2.26 0.159 11 = 0.511
b=

and

352.9 12 11 2.26

Y = 0.511 + 0.159X
(c)Given a tourist population of 10,000,000, the model
predicts a ridership of:
Y = 0.511 + 0.159 10 = 2.101, or 2,101,000 persons.
(d)If there are no tourists at all, the model predicts a
ridership of 0.511, or 511,000 persons. One would not
place much confidence in this forecast, however,
because the number of tourists (zero) is outside the
range of data used to develop the model.
(e)The standard error of the estimate is given by:
Syx =

Therefore:

XY nXY

a = Y bX

a = Y bX

b=

Ridership
(1,000,000s)
(Y)

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

and X = 854, Y = 478, XY = 42558.6, X2 = 76129.9,


Y 2 = 23892, X = 85.4, Y = 47.8. Then:

Tourists
(Millions)
(X)

Y 2 a Y b XY
n2
71.59 0.511 27.1 0.159 352.9
12 2

71.59 13.85 56.11


= .163
10
= .404 (rounded to .407 in POM for Windows software)

Copyright 2011 Pearson Education, Inc. publishing=as Prentice Hall.

CHAPTER 4F O R E C A S T I N G

(f) The correlation coefficient and the coefficient of determination are given by:
r=

=
=

n XY X Y
n X 2 ( X ) 2 n Y 2 ( Y ) 2

12 352.9 132 27.1

Apr.
May.
Jun.

35
42
50

Oct.
Nov.
Dec.

41

35
38
29

Both history and forecast for the next year are shown in the
accompanying figure:

12 1796 1322 12 71.59 27.12

4234.8 3577.2
21552 17424 859.08 734.41
657.6
657.6
=
= 0.917
4128 124.67 64.25 11.166

and r 2 = 0.840
4.42(a)This problem gives students a chance to tackle a
realistic problem in business, i.e., not enough data to
make a good forecast. As can be seen in the
accompanying figure, the data contains both seasonal
and trend factors.

4.43(a) and (b) See the following table:


Week
t

Averaging methods are not appropriate with trend, seasonal,


or other patterns in the data. Moving averages smooth out seasonality. Exponential smoothing can forecast January next year,
but not farther. Because seasonality is strong, a nave model that
students create on their own might be best.
(b) One model might be: Ft+1 = At11
That is forecastnext period = actualone year earlier to account
for
seasonality. But this ignores the trend.
One very good approach would be to calculate the increase
from each month last year to each month this year, sum all 12
increases, and divide by 12. The forecast for next year would
equal the value for the same month this year plus the average
increase over the 12 months of last year.
(c) Using this model, the January forecast for next year becomes:
F25 = 17 +

148
= 17 + 12 = 29
12

where 148 = total monthly increases from last year to this year.
The forecasts for each of the months of next year then become:
Jan.
Feb.
Mar.

29
26
32

July.
Aug.
Sep.

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
A(t)

Smoothed
Value
Ft ( = 0.2)

50
35
25
40
45
35
20
30
35
20
15
40
55
35
25
55
55
40
35
60
75
50
40
65

+50.0
+50.0
+47.0
+42.6
+42.1
+42.7
+41.1
+36.9
+35.5
+35.4
+32.3
+28.9
+31.1
+35.9
+36.7
+33.6
+37.8
+41.3
+41.0
+39.8
+43.9
+50.1
+50.1
+48.1
+51.4

Forecast
Error
+0.0
15.0
22.0
2.6
2.9
7.7
21.1
6.9
0.5
15.4
17.3
+11.1
+23.9
0.9
10.7
+21.4
+17.2
1.3
6.0
+20.2
+31.1
0.1
10.1
+16.9

MAD = 11.8

Smoothed
Value
Forecast
Ft ( = 0.6)
Error
+50.0
+50.0
+41.0
+31.4
+36.6
+41.6
+37.6
+27.1
+28.8
+32.5
+25.0
+19.0
+31.6
+45.6
+39.3
+30.7
+45.3
+51.1
+44.4
+38.8
+51.5
+65.6
+56.2
+46.5
+57.6

+0.0
15.0
16.0
+8.6
+8.4
6.6
17.6
+2.9
+6.2
12.5
10.0
+21.0
+23.4
10.6
14.3
+24.3
+9.7
11.1
9.4
+21.2
+23.5
15.6
16.2
+18.5
MAD = 13.45

(c)Students should note how stable the smoothed values


are for = 0.2. When compared to actual week 25 calls
of 85, the smoothing constant, = 0.6, appears to do a
slightly better job. On the basis of the standard error
of the estimate and the MAD, the 0.2 constant is
better. However, other smoothing constants need to be
examined.

56
53
45

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

42

CHAPTER 4F O R E C A S T I N G

4.44
Week
t

Actual Value
At

Smoothed Value
Ft ( = 0.3)

Trend Estimate
Tt ( = 0.2)

Forecast
FITt

Forecast
Error

50.000
35.000
25.000
40.000
45.000
35.000
20.000
30.000
35.000
20.000
15.000
40.000
55.000
35.000
25.000
55.000
55.000
40.000
35.000
60.000
75.000
50.000
40.000
65.000

50.000
50.000
45.500
38.720
37.651
38.543
36.555
30.571
28.747
29.046
25.112
20.552
24.526
32.737
33.820
31.649
38.731
44.664
44.937
43.332
49.209
58.470
58.445
54.920
59.058

0.000
0.000
0.900
2.076
1.875
1.321
1.455
2.361
2.253
1.743
2.181
2.657
1.331
0.578
0.679
0.109
1.503
2.389
1.966
1.252
2.177
3.594
2.870
1.591
2.100

50.000
50.000
44.600
36.644
35.776
37.222
35.101
28.210
26.494
27.303
22.931
17.895
23.196
33.315
34.499
31.758
40.234
47.053
46.903
44.584
51.386
62.064
61.315
56.511
61.158

0.000
15.000
19.600
3.356
9.224
2.222
15.101
1.790
8.506
7.303
7.931
22.105
31.804
1.685
9.499
23.242
14.766
7.053
11.903
15.416
23.614
12.064
21.315
8.489

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

To evaluate the trend adjusted exponential smoothing model,


actual week 25 calls are compared to the forecasted value. The
model appears to be producing a forecast approximately midrange between that given by simple exponential smoothing using
= 0.2 and = 0.6. Trend adjustment does not appear to give
any significant improvement.
n

Tracking signal =

4.45

( At Ft )

t =1

Month

At

Ft

May
June
July
August
September
October
November
December

100
80
110
115
105
110
125
120

100
104
99
101
104
104
105
109

MAD
|At Ft |
0
24
11
14
1
6
20
11
Sum: 87

(At Ft )
0
24
11
14
1
6
20
11
Sum: 39

87
= 10.875
8
39
= 3.586
10.875

So: MAD:

4.46 (a)

421
377

2.90
2.93

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

X2
177241
142129

Y2
8.41
8.58

XY
1220.9
1104.6

CHAPTER 4F O R E C A S T I N G

Column totals

585
690
608
390
415
481
729
501
613
709
366
6885

3.00
3.45
3.66
2.88
2.15
2.53
3.22
1.99
2.75
3.90
1.60
36.96

342225
476100
369664
152100
172225
231361
531441
251001
375769
502681
133956
3857893

9.00
11.90
13.40
8.29
4.62
6.40
10.37
3.96
7.56
15.21
2.56
110.26

1755.0
2380.5
2225.3
1123.2
892.3
1216.9
2347.4
997.0
1685.8
2765.1
585.6
20299.5

Given: Y = a + bX where:
b=

XY nXY
X 2 nX 2

a = Y bX

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

43

44

CHAPTER 4F O R E C A S T I N G

and X = 6885, Y = 36.96, XY = 20299.5, X 2 = 3857893,


Y 2 = 110.26, X = 529.6, Y = 2.843, Then:
b=

20299.5 13 529.6 2.843

3857893 13 529.62
726
=
= 0.0034
211703
a = 2.84 0.0034 529.6 = 1.03

20299.5 19573.5
3857893 3646190

Amits MAD for exponential smoothing (16.11) is lower


than that of Barbaras moving average (19.17). So his
forecast seems to be better.

and Y = 1.03 + 0.0034X


As an indication of the usefulness of this relationship, we can
calculate the correlation coefficient:
r=

=
=

n XY X Y
n X 2 ( X ) 2 n Y 2 ( Y ) 2

13 20299.5 6885 36.96


13 3857893 68852 13 110.26 36.962

263893.5 254469.6
50152609 47403225 1433.4 1366.0
9423.9

2749384 67.0
9423.9
=
= 0.692
1658.13 8.21
r 2 = 0.479
A correlation coefficient of 0.692 is not particularly high. The
coefficient of determination, r2, indicates that the model explains
only 47.9% of the overall variation. Therefore, while the model
does provide an estimate of GPA, there is considerable variation
in GPA, which is as yet unexplained. For
(b) X = 350: Y = 1.03 + 0.0034 350 = 2.22
(c) X = 800: Y = 1.03 + 0.0034 800 = 3.75
Note: When solving this problem, care must be taken to interpret
significant digits. Also note that X = 800 is outside the range of the
data set used to determine the regression relationship, so caution is
advised.
4.47(a)There is not a strong linear trend in sales over time.
(b, c)Amit wants to forecast by exponential smoothing (setting
Februarys forecast equal to Januarys sales) with alpha =
0.1. Barbara wants to use a 3-period moving average.
January
February
March
April
May

Sales

Amit

Barbara Amit Error

400
380
410
375
405

400
398
399.2
396.8

396.67
388.33
MAD =

20.0
12.0
24.2
8.22
16.11

Barbara Error

21.67
16.67
19.17

(d) Note that Amit has more forecast observations, while


Barbaras moving average does not start until month 4. Also
note that the MAD for Amit is an average of 4 numbers,
while Barbaras is only 2.

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

CHAPTER 4F O R E C A S T I N G

4.48(a)

45

4.49(a) (Continued)

Quarter

Contracts X

1
2
3
4
5
6
7
8
Totals
Average

153
172
197
178
185
199
205
226
1,515
189.375

Sales Y

8
10
15
9
12
13
12
16
95
11.875

23,409
29,584
38,809
31,684
34,225
39,601
42,025
51,076
290,413

64
100
225
81
144
169
144
256
1,183

Method Exponential Smoothing


0.6 =
Deposits (Y )
Forecast
|Error|

XY
1,224
1,720
2,955
1,602
2,220
2,587
2,460
3,616
18,384

Year

30
31
32
33
34
35
36
37
38
39
b = (18384 8 189.375 11.875)/(290,413 8 189.375
40
189.375) = 0.1121
41
a = 11.875 0.1121 189.375 = 9.3495
42
Sales ( y) = 9.349 + 0.1121 (Contracts)
43
44
(b)
TOTALS
r = (8 18384 1515 95) ((8 290,413 15152 )(8 1183 952AVERAGE
))

= 0.8963
Sxy = 1183 ( 9.3495 95) (0.112 18384 / 6) = 1.3408

Year

Year

Method Exponential Smoothing


0.6 =
Deposits (Y ) Forecast
|Error|

Error2

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
29

0.25
0.24
0.24
0.26
0.25
0.30
0.31
0.32
0.24
0.26
0.25
0.33
0.50
0.95
1.70
2.30
2.80
2.80
2.70
3.90
4.90
5.30
6.20
4.10
4.50
6.10
7.70
10.10
15.20

0.00
0.0001
0.0000
0.0003
0.00
0.0023
0.0008
0.0004
0.0051
0.0000
0.0002
0.0055
0.0399
0.2808
0.925
0.9698
0.7990
0.1278
0.0018
1.4816
2.2108
0.9895
1.6845
2.499
0.0540
2.2712
4.8524
10.7658
41.1195

0.25
0.25
0.244
0.241
0.252
0.251
0.280
0.298
0.311
0.268
0.263
0.255
0.300
0.420
0.738
1.315
1.906
2.442
2.656
2.682
3.413
4.305
4.90
5.680
4.732
4.592
5.497
6.818
8.787

0.00
0.01
0.004
0.018
0.002
0.048
0.029
0.021
0.071
0.008
0.013
0.074
0.199
0.529
0.961
0.984
0.893
0.357
0.043
1.217
1.486
0.994
1.297
1.580
0.232
1.507
2.202
3.281
6.412

(Continued)

12.6350
15.9140
20.8256
23.69
27.6561
32.6624
31.72
31.71
35.784
43.0536
46.6814
52.1526
62.9210
67.7084
74.5434

Next period forecast = 86.2173

r 2 = .8034
4.49(a)

18.10
24.10
25.60
30.30
36.00
31.10
31.70
38.50
47.90
49.10
55.80
70.10
70.90
79.10
94.00
787.30
17.8932

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
29
30
31
32
33
34

Error2

5.46498
29.8660
8.19
67.01
4.774
22.7949
6.60976
43.69
8.34390
69.62
1.56244
2.44121
0.024975 0.000624
6.79
46.1042
12.116
146.798
6.046
36.56
9.11856
83.1481
17.9474
322.11
7.97897
63.66
11.3916
129.768
19.4566
378.561
150.3
1,513.22
3.416
34.39
(MAD)
(MSE)
Standard error = 6.07519

Method Linear Regression (Trend Analysis)


Period (X)
Deposits (Y )
Forecast
Error2
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
29
30
31
32
33
34

0.25
0.24
0.24
0.26
0.25
0.30
0.31
0.32
0.24
0.26
0.25
0.33
0.50
0.95
1.70
2.30
2.80
2.80
2.70
3.90
4.90
5.30
6.20
4.10
4.50
6.10
7.70
10.10
15.20
18.10
24.10
25.60
30.30
36.00

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

17.330
15.692
14.054
12.415
10.777
9.1387
7.50
5.8621
4.2238
2.5855
0.947
0.691098
2.329
3.96769
5.60598
7.24427
8.88257
10.52
12.1592
13.7974
15.4357
17.0740
18.7123
20.35
21.99
23.6272
25.2655
26.9038
28.5421
30.18
31.8187
33.46
35.0953
36.7336

309.061
253.823
204.31
160.662
121.594
89.0883
61.0019
38.2181
19.9254
8.09681
1.43328
0.130392
3.34667
9.10642
15.2567
24.4458
36.9976
59.6117
89.4756
97.9594
111.0
138.628
156.558
264.083
305.862
307.203
308.547
282.367
178.011
145.936
59.58
61.73
22.9945
0.5381

46

CHAPTER 4F O R E C A S T I N G

Forecasting Summary Table


Method used:

MAD
MSE
Standard error using
n 2 in denominator
Correlation coefficient
35
36
37
38
39
40
41
42
43
44
TOTALS
AVERAGE

35
36
37
38
39
40
41
42
43
44

31.10
31.70
38.50
47.90
49.10
55.80
70.10
70.90
79.10
94.00

990.00

787.30

22.50

17.893

38.3718
40.01
41.6484
43.2867
44.9250
46.5633
48.2016
49.84
51.4782
53.1165

52.8798
69.0585
9.91266
21.2823
17.43
85.3163
479.54
443.528
762.964
1,671.46
7,559.
95
171.817
(MSE)

Method Least squaresSimple Regression on GSP


a
b
17.636
13.5936
Coefficients:
GSP Deposits
Year
(X)
(Y ) Forecast
|Error|
Error2
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

0.40
0.40
0.50
0.70
0.90
1.00
1.40
1.70
1.30
1.20
1.10
0.90
1.20
1.20
1.20
1.60
1.50
1.60
1.70
1.90
1.90
2.30
2.50
2.80
2.90
3.40

0.25
0.24
0.24
0.26
0.25
0.30
0.31
0.32
0.24
0.26
0.25
0.33
0.50
0.95
1.70
2.30
2.80
2.80
2.70
3.90
4.90
5.30
6.20
4.10
4.50
6.10

12.198
12.4482
12.198
12.4382
10.839
11.0788
8.12
8.38
5.4014 5.65137
4.0420 4.342
1.39545 1.08545
5.47354 5.15354
0.036086 0.203914
1.3233 1.58328
2.6826 2.93264
5.4014 5.73137
1.3233 1.82328
1.3233 2.27328
1.3233 3.02328
4.11418 1.81418
2.75481 0.045186
4.11418 1.31418
5.47354 2.77354
8.19227 4.29227
8.19227 3.29227
13.6297 8.32972
16.3484 10.1484
20.4265 16.3265
21.79
17.29
28.5827 22.4827

154.957
154.71
122.740
70.226
31.94
18.8530
1.17820
26.56
0.041581
2.50676
8.60038
32.8486
3.32434
5.16779
9.14020
3.29124
0.002042
1.727
7.69253
18.4236
10.8390
69.3843
102.991
266.556
298.80
505.473

Exponential
Smoothing

Linear Regression
(Trend Analysis)

Linear Regression

3.416
34.39
6.075

Y = 18.968 +
1.638 YEAR
10.587
171.817
13.416

Y = 17.636 +
13.59364 GSP
10.255
204.919
14.651

0.846

0.813

27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
TOTALS
AVERAGE

3.80
4.10
4.00
4.00
3.90
3.80
3.80
3.70
4.10
4.10
4.00
4.50
4.60
4.50
4.60
4.60
4.70
5.00

7.70
10.10
15.20
18.10
24.10
25.60
30.30
36.00
31.10
31.70
38.50
47.90
49.10
55.80
70.10
70.90
79.10
94.00

34.02
38.0983
36.74
36.74
35.3795
34.02
34.02
32.66
38.0983
38.0983
36.74
43.5357
44.8951
43.5357
44.8951
44.8951
46.2544
50.3325

26.32
27.9983
21.54
18.64
11.2795
8.42018
3.72018
3.33918
6.99827
6.39827
1.76
4.36428
4.20491
12.2643
25.20
26.00
32.8456
43.6675
451.223
10.2551
(MAD)

692.752
783.90
463.924
347.41
127.228
70.8994
13.8397
11.15
48.9757
40.9378
3.10146
19.05
17.6813
150.412
635.288
676.256
1,078.83
1,906.85
9,016.45
204.92
(MSE)

Given that one wishes to develop a five-year forecast,


trend analysis is the appropriate choice. Measures of
error and goodness-of-fit are really irrelevant.
Exponential smoothing provides a forecast only of
deposits for the next yearand thus does not address
the five-year forecast problem. In order to use the
regression model based upon GSP, one must first
develop a model to forecast GSP, and then use the
forecast of GSP in the model to forecast deposits. This
requires the development of two modelsone of which
(the model for GSP) must be based solely on time as the
independent variable (time is the only other variable we
are given).
(b)One could make a case for exclusion of the older data.
Were we to exclude data from roughly the first 25 years,
the forecasts for the later years would likely be
considerably more accurate. Our argument would be that
a change that caused an increase in the rate of growth
appears to have taken place at the end of that period.
Exclusion of this data, however, would not change our
choice of forecasting model because we still need to
forecast deposits for a future five-year period.

ADDITIONAL HOMEWORK PROBLEMS

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

CHAPTER 4F O R E C A S T I N G

These problems, which appear on www.myomlab.com, provide an


additional 13 problems that you may wish to assign.
4.50

(a)
(b)
(c)

Week

10

Forecast

Registration
Nave
2-week moving
4-week moving

22

21
22

25
21
21.5

27
25
23

35
27
26
23.75

29
35
31
27

33
29
32
29

37
33
31
31

41
37
35
33.5

37
41
39
35

37
39
37

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

47

48

CHAPTER 4F O R E C A S T I N G

4.51

(c)Based on
a Mean Absolute
Deviation
criterion,Absolute
the
Sales
Three-Month
Moving Average
Moving
Deviation

Month
January
Exponentially Smoothed Forecast
February
7
5
March
9
5 + 0.2 (7 5) = 5.4
April
5
5.4 + 0.2 (9 5.4) = 6.12
May
9
6.12 + 0.2 (5 6.12) = 5.90
June
13
5.90 + 0.2 (9 5.90) = 6.52
July
8
6.52 + 0.2 (13 6.52) = 7.82
August
Forecast
7.82 + 0.2 (8 7.82) = 7.86
September
October
November
Forecast
|Error|
December
Error2
January
100
5
25
February
110
2
4
120
3
9
130
0
0
10
38

Period

Demand

1
2
3
4
5
6
7

4.52
Actual
95
108
123
130

MAD = 10/4 = 2.5, MSE = 38/4 = 9.5


4.53(a)3-month moving average:
Month

Sales

January
February
March
April
May
June
July
August
September
October
November
December
January
February

11
14
16
10
15
17
11
14
17
12
14
16
11

Three-Month
Moving Average

Absolute
Deviation

(11 + 14 + 16)/3 = 13.67


(14 + 16 + 10)/3 = 13.33
(16 + 10 + 15)/3 = 13.67
(10 + 15 + 17)/3 = 14.00
(15 + 17 + 11)/3 = 14.33
(17 + 11 + 14)/3 = 14.00
(11 + 14 + 17)/3 = 14.00
(14 + 17 + 12)/3 = 14.33
(17 + 12 + 14)/3 = 14.33
(12 + 14 + 16)/3 = 14.00
(14 + 16 + 11)/3 = 13.67

3.67
1.67
3.33
3.00
0.33
3.00
2.00
0.33
1.67
3.00

= 22.00
MAD = 2.20
(b)3-month weighted moving average

3-month moving average with MAD = 2.2 is to be

11
preferred over the 3-month weighted moving average
14
with MAD = 2.72.
16
(d)Other
factors
a more complex
10
(1 that
11 +might
2 14be
+ included
3 16)/6 =in14.50
15 model are interest
(1 14 +rates
2 16
+ 3cycle
10)/6
12.67 factors.
and
or=seasonal
17
(1 16 + 2 10 + 3 15)/6 = 13.50
4.54(a)
11
(1 10 + 2 15 + 3 17)/6 = 15.17
14
(1 15 + 2Cumulative
17 + 3 11)/6 = 13.67
Actual
Cum. Tracking
17
17 + 2 Error
11 + 3 14)/6
= 13.50
Week Miles
Forecast(1 Error
|Error|
MAD Signal
1217
11 + 2 14 + 3 17)/6
= 15.00
1
17.00 (1 0.00
0.00
0
1421
14 + 2 17
+ 3 12)/6
= 14.00
2
17.00 (1 4.00
4.00
4.00
2
2
1619
17 + 2 12
+ 3 14)/6
= 13.831.73 3
3
17.80 (1 1.20
5.20
5.20
1123
12 + 2 10.16
14 + 3 16)/6
= 14.672.54 4
4
18.04 (1 4.96
10.16
14 + 2 16
+ 3 11)/6
= 13.172.24 4
5
18
19.03 (1 +1.03
9.13
11.19
6
7
8
9
10
11
12

16
20
18
22
20
15
22

18.83
18.26
18.61
18.49
19.19
19.35
18.48

+2.83
1.74
+0.61
3.51
0.81
+4.35
3.52

6.30
8.04
7.43
10.94
11.75
7.40
10.92

14.02
15.76
16.37
19.88
20.69
25.04
28.56

2.34
2.7
= 27.17
2.25
3.6
MAD = 2.72
2.05
3.6
2.21 5
2.07
5.7
2.28
3.2
2.38
4.6

(b)The MAD = 28.56/12 = 2.38


(c)The cumulative error and tracking signals appear to
be consistently negative, and at week 10, the tracking
signal exceeds 5 MADs.
4.55

x2

xy

7
9
5
11
10
13
55

1
2
3
4
5
6
21

1
4
9
16
25
36
91

7
18
15
44
50
78
212

y = 9.17
x = 3.5
y = 5.27 + 1.11x
Period 7 forecast = 13.07
Period 12 forecast = 18.64, but this is far outside the range
of valid data.

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

4.50
2.33
3.50
4.17
0.33
3.50
3.00
0.00
2.17
3.67

CHAPTER 4F O R E C A S T I N G

4.56To compute seasonalized or adjusted sales forecast, we just


multiply each seasonalized index by the appropriate trend forecast.
Y
= Index Y
Seasonal

III

Quarter IV: YIV = 1.10 180,000 = 198,000


4.57
Mon

Tue

Wed

178
180
176
178
178

250
250
260
260
255

Thu
215
213
220
225
218.3

n X ( X ) 2 n Y 2 ( Y ) 2

5 70 15 20

5 55 152 5 130 202

350 300
50
=
=
50 250
275 225 650 400
50
=
= 0.45
111.80

Quarter I: YI = 1.25 120,000 = 150,000


Quarter II: YII = 0.90 140,000 = 126,000
Quarter III: Y = 0.75 160,000 = 120,000

210
215
220
225
217.5

n XY X Y
2

Trend forecast

Hence, for

Week 1
Week 2
Week 3
Week 4
Averages

r=

49

Fri

Sat

160
165
175
176
169

180
185
190
190
186.3

Overall average = 204

(a)Seasonal indices:
1.066 (Mon)0.873 (Tue)1.25 (Wed)
1.07 (Thu)0.828 (Fri)0.913 (Sat)

(b)To calculate for Monday of Week 5 = 201.74 +


0.18(25) 1.066 = 219.9 rounded to 220
Forecast

The correlation coefficient indicates that there is a positive


correlation between bank deposits and consumer price indices in

220 (Mon) 180 (Tue) 258 (Wed)


221 (Thu) 171 (Fri)189 (Sat)

4.58(a)4000 + 0.20(15,000) = 7,000


(b)4000 + 0.20(25,000) = 9,000
4.59(a)35 + 20(80) + 50(3.0) = 1,785
(b)35 + 20(70) + 50(2.5) = 1,560
4.60Given: X = 15, Y = 20, XY = 70, X2 = 55, Y2 = 130,
X = 3, Y = 4
XY nXY
(a)
b=
X 2 nX 2
a = Y bX
70 5 3 4 70 60 10
b=
=
=
=1
55 45 10
55 5 32
a = 4 1 3 = 4 3 = 1
Y = 1 + 1X

Birmingham, Alabamai.e., as one variable tends to increase (or


decrease), the other tends to increase (or decrease).
4.60(c)Standard error of the estimate:

(b) Correlation coefficient:

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

50

CHAPTER 4F O R E C A S T I N G

Syx =
=

Y 2 a Y b XY
130 1 20 1 70
=
n2
3
130 20 70
=
3

40
= 13.3 = 3.65
3

4.61

Column Totals

X2

Y2

XY

2
1
4
5
3

4
1
4
6
5

4
1
16
25
9

16
1
16
36
25

8
1
16
30
15

15

20

55

94

70

Given: Y = a + bX where:
b=

XY nXY
X 2 nX 2

a = Y bX
and X = 15, Y = 20, XY = 70, X2 = 55, Y2 = 94, X = 3,
Y = 4. Then:
b=

70 5 4 3

55 5 3
a = 4 1 3 = 1.0
2

70 60
= 1.0
55 45

and Y = 1.0 + 1.0X. The correlation coefficient:


r=

n XY X Y
n X 2 ( X ) 2 n Y 2 ( Y ) 2

5 70 15 20
5 55 152 5 94 202

50
50
=
= 0.845
50 70 59.16

350 300
275 225 470 400

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

CHAPTER 4F O R E C A S T I N G

11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36

487
525
575
527
540
502
508
573
508
498
485
526
552
587
608
597
612
603
628
605
627
578
585
581
632
656

121
144
169
196
225
256
289
324
361
400
441
484
529
576
625
676
729
784
841
900
961
1,024
1,089
1,156
1,225
1,296

5,357
6,300
7,475
7,378
8,100
8,032
8,636
10,314
9,652
9,960
10,185
11,572
12,696
14,088
15,200
15,522
16,524
16,884
18,212
18,150
19,437
18,496
19,305
19,754
22,120
23,616

237,169
275,625
330,625
277,729
291,600
252,004
258,064
328,329
258,064
248,004
235,225
276,676
304,704
344,569
369,664
356,409
374,544
363,609
394,384
366,025
393,129
334,084
342,225
337,561
399,424
430,336

666

19,366

16,206

378,661

10,558,246

Standard error of the estimate:


Syx =
=

Y 2 a Y b XY
=
n2

94 1 20 1 70
52

94 20 70
= 1.333 = 1.15
3

4.62Using software, the regression equation is: Games lost =


6.41 + 0.533 days rain.

CASE STUDIES
SOUTHWESTERN UNIVERSITY: B

This is the second in a series of integrated case studies that run


throughout the text.
1.One way to address the case is with separate forecasting models
for each game. Clearly, the homecoming game (week 2) and the
fourth game (craft festival) are unique attendance situations.
Forecasts
2010
2011

Game

Model

1
2
3
4
5
Total

y = 30,713 + 2,534x
y = 37,640 + 2,146x
y = 36,940 + 1,560x
y = 22,567 + 2,143x
y = 30,440 + 3,146x

48,453
52,660
47,860
37,567
52,460
239,000

50,988
54,806
49,420
39,710
55,606
250,530

R2
0.92
0.90
0.91
0.88
0.93

(where y = attendance and x = time)


2.Revenue in 2010 = (239,000) ($50/ticket) = $11,950,000
Revenue in 2011 = (250,530) ($52.50/ticket) = $13,152,825

Totals
Average

b=

2 DIGITAL CELL PHONE, INC.


Objectives:

Selection of an appropriate time series forecasting model based


upon a plot of the data.

The importance of combining a qualitative model with a


quantitative model in situations where technological change is
occurring.

1.A plot of the data indicates a linear trend (least squares)


model might be appropriate for forecasting. Using linear trend
you obtain the following:
x

1
2
3
4
5
6
7
8
9
10

480
436
482
448
458
489
498
430
444
496

x2
1
4
9
16
25
36
49
64
81
100

xy
480
872
1,446
1,792
2,290
2,934
3,486
3,440
3,996
4,960

y2
230,400
190,096
232,324
200,704
209,464
239,121
248,004
184,900
197,136
246,016

537.9

450.2

10,518.4

378,661 36 18.5 537.9

x nx
16,206 (36 18.5 )
a = y bx = 537.9 5.25 18.5 = 440.85

3.In games 2 and 5, the forecast for 2011 exceeds stadium


capacity. With this appearing to be a continuing trend, the time
has come for a new or expanded stadium.

18.5

xy nx y

r=

51

293,284.6

20390.0
= 5.25
3885.0

n xy x y
[ n x ( x )2 ][ n y2 ( y)2 ]
2

(36)(378,661) (666)(19,366)
[(36) (16,206) (666)2 ][(36)(10,558,246) (19,366)2 ]
13,631,796 12,897,756
[(583,416) (443,556)][380,096,856) (375,041,956)]
737,040
[139,860][5,054,900]

734,040
706,978,314,000

734,040
= .873
840,820

r 2 = .76
y = 440.85 + 5.25 (time)
r = 0.873 indicating a reasonably good fit
The student should report the linear trend results, but deflate
the forecast obtained based upon qualitative information about
industry and technology trends.
Because there is limited seasonality in the data, the linear
trend analysis above provides a good r2 of .76.
However, a more precise forecast can be developed addressing
the seasonality issue, which is done below. Methods a and c yield r2
of .85 and .86, respectively, and methods b and d, which also center
the seasonal adjustment, yield r2 of .93 and .94, respectively.
2.Four approaches to decomposition of The Digital Cell Phone
data can address seasonality, as follows:

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

52

CHAPTER 4F O R E C A S T I N G

At $65,000; y = 8.3 + .8 (65) = 8.3 + 52 = 60.3, or 60,300 guests.

a) Multiplicative seasonal model,


Cases = 443.87 + 5.08 (time), r2 = .85, MAD = 20.89
b) Multiplicative Seasonal Model, with centered moving averages
(CMA), which is not covered in our text but can be seen in
Render, Stair, and Hannas Quantitative Analysis for
Management, 10th ed., Prentice Hall Publishing.
Cases = 432.28 + 5.73 (time), r2 = .93, MAD = 12.84
c) Additive seasonal model,

For the instructor who asks other questions than this one:
r2 = 0.8869
Std. error = 5.062

ADDITIONAL CASE STUDY*


THE NORTH-SOUTH AIRLINE

Cases = 444.29 + 5.06 (time), r2 = .86, MAD = 20.02

Northern Airline Data

d) Additive seasonal model, with centered moving averages.


Cases = 431.31 + 5.72 (time), r2 = .94, MAD = 12.28
The two methods that use the average of all data have very
similar results, and the two CMA methods also look quite close. As
suggested with this analysis, CMA is typically the better technique.

VIDEO CASE STUDY


FORECASTING AT HARD ROCK CAFE

Year

Airframe Cost
per Aircraft

Engine Cost
per Aircraft

Average
Age (hrs)

2003
2004
2005
2006
2007
2008
2009

51.80
54.92
69.70
68.90
63.72
84.73
78.74

43.49
38.58
51.48
58.72
45.47
50.26
79.60

6512
8404
11077
11717
13275
15215
18390

Year

Airframe Cost
per Aircraft

Engine Cost
per Aircraft

Average
Age (hrs)

2003
2004
2005
2006
2007
2008
2009

13.29
25.15
32.18
31.78
25.34
32.78
35.56

18.86
31.55
40.43
22.10
19.69
32.58
38.07

5107
8145
7360
5773
7150
9364
8259

There is a short video (8 minutes) available from Prentice Hall


and filmed specifically for this text that supplements this case.
1.Hard Rock uses forecasting for (1) sales (guest counts) at cafes,
(2) retail sales, (3) banquet sales, (4) concert sales, (5) evaluating
managers, and (6) menu planning. They could also employ these
techniques to forecast: retail store sales of individual (SKU) product
demands; sales of each entre; sales at each work station, etc.
2.The POS system captures all the basic sales data needed to
drive individual cafes scheduling/ordering. It also is aggregated
at corporate HQ. Each entre sold is counted as one guest at a
Hard Rock Cafe.
3.The weighting system is subjective, but is reasonable. More
weight is given to each of the past 2 years than to 3 years ago.
This system actually protects managers from large sales variations
outside their control. One could also justify a 50%30%20%
model or some other variation.
4.Other predictors of cafe sales could include season of year
(weather); hotel occupancy; spring break from colleges; beef
prices; promotional budget; etc.
5.Y = a + bx
Month
1
2
3
4
5
6
7
8
9
10
Totals
Average

b=

Advertising X
14
17
25
25
35
35
45
50
60
60
366
36.6

Guest Count Y
21
24
27
32
29
37
43
43
54
66
376
37.6

15,772 10 36.6 37.6

X2

Y2

196
441
289
576
625
729
625 1,024
1,225
841
1,225 1,369
2,025 1,849
2,500 1,849
3,600 2,916
3,600 4,356
15,910 15,950

XY
294
408
675
800
1,015
1,295
1,935
2,150
3,240
3,960
15,772

Southeast Airline Data

Utilizing the software package provided with this text, we


can develop the following regression equations for the variables
of interest:
Northern AirlinesAirframe Maintenance Cost:
Cost = 36.10 + 0.0026 Airframe age
Coefficient of determination = 0.7695
Coefficient of correlation = 0.8772
Northern AirlinesEngine Maintenance Cost:
Cost = 20.57 + 0.0026 Airframe age
Coefficient of determination = 0.6124
Coefficient of correlation = 0.7825
Southeast AirlinesAirframe Maintenance Cost:
Cost = 4.60 + 0.0032 Airframe age
Coefficient of determination = 0.3905
Coefficient of correlation = 0.6249
Southeast AirlinesEngine Maintenance Cost;
Cost = 0.67 + 0.0041 Airframe age
Coefficient of determination = 0.4600
Coefficient of correlation = 0.6782
The following graphs portray both the actual data and the
regression lines for airframe and engine maintenance costs for
both airlines.

= 0.7996 .8
15,910 10 36.62
a = 37.6 0.7996 36.6 = 8.3363 8.3
Y = 8.3363 + 0.7996 X
Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.

CHAPTER 4F O R E C A S T I N G

*This case study appears on our companion Web site, at


www.pearsonhighered.com/heizer, and at www.myomlab.com.

53

Southeast Airlines: The relationships between maintenance costs and airframe age for Southeast Airlines are
much less well defined. It is even more obvious that
airframe age is not the only important factorperhaps not
even the most important factor.

Overall, it would seem that:

Northern Airlines has the smallest variance in maintenance costsindicating that its day-to-day management
of maintenance is working pretty well.

Maintenance costs seem to be more a function of airline


than of airframe age.

The airframe and engine maintenance costs for Southeast


Airlines are not only lower, but more nearly similar than
those for Northern Airlines. From the graphs, at least, they
appear to be rising more sharply with age.

From an overall perspective, it appears that Southeast Airlines


may perform more efficiently on sporadic or emergency
repairs, and Northern Airlines may place more emphasis
on preventive maintenance.

Ms. Youngs report should conclude that:

There is evidence to suggest that maintenance costs could


be made to be a function of airframe age by implementing
more effective management practices.

The difference between maintenance procedures of the two


airlines should be investigated.
The data with which she is currently working does not
provide conclusive results.

Note that the two graphs have been drawn to the same scale
to facilitate comparisons between the two airlines.

Concluding Comment:
The question always arises, with this case, as to whether the data
should be merged for the two airlines, resulting in two regressions
instead of four. The solution provided is that of the consultant
who was hired to analyze the data. The airlines own internal
analysts also conducted regressions, but did merge the data sets.
This shows how statisticians can take different views of the same
data.

Comparison:

Northern Airlines: There seem to be modest correlations


between maintenance costs and airframe age for Northern
Airlines. There is certainly reason to conclude, however,
that airframe age is not the only important factor.

Copyright 2011 Pearson Education, Inc. publishing as Prentice Hall.