SCMA 311

Chapter 4: Forecasting

Section 1: What is Forecasting?

1) Forecasts may be influenced by a product's position in its life cycle.

Answer: TRUE

2) Demand forecasts serve as inputs to financial, marketing, and personnel planning.

Answer: TRUE

3) What two numbers are contained in the daily report to the CEO of Walt Disney Parks & Resorts regarding the six
Orlando parks?
A) yesterday's forecasted attendance and yesterday's actual attendance
B) yesterday's actual attendance and today's forecasted attendance
C) yesterday's forecasted attendance and today's forecasted attendance
D) yesterday's actual attendance and last year's actual attendance
E) yesterday's forecasted attendance and the year-to-date average daily forecast error

4) As compared to long-range forecasts, short-range forecasts:

A) are less accurate.
B) deal with less comprehensive issues supporting management decisions.
C) employ similar methodologies.
D) all of the above
E) none of the above

5) One use of short-range forecasts is to determine:

A) planning for new products.
B) capital expenditures.
C) research and development plans.
D) facility location.
E) job assignments.

6) Forecasts are usually classified by time horizon into which three categories?
A) short-range, medium-range, and long-range
B) finance/accounting, marketing, and operations
C) strategic, tactical, and operational
D) exponential smoothing, regression, and time series
E) departmental, organizational, and industrial

7) A forecast with a time horizon of about 3 months to 3 years is typically called a:

A) long-range forecast.
B) medium-range forecast.
C) short-range forecast.
D) weather forecast.
E) strategic forecast.

8) Forecasts used for new product planning, capital expenditures, facility location or expansion, and R&D typically
utilize a:
A) short-range time horizon.
B) medium-range time horizon.
C) long-range time horizon.
D) naive method, because there is no data history.
E) trend extrapolation.

4-1
9) The three major types of forecasts used by organizations in planning future operations are:
A) strategic, tactical, and operational.
B) economic, technological, and demand.
C) exponential smoothing, Delphi, and regression.
D) causal, time-series, and seasonal.
E) departmental, organizational, and territorial.

10) Which of the following most requires long-range forecasting (as opposed to short-range or medium-range
forecasting) for its planning purposes?
A) job scheduling
B) production levels
C) cash budgeting
D) capital expenditures
E) purchasing

11) Short-range forecasts tends to ____more____ longer-range forecasts.

A) be less accurate than
B) be more accurate than
C) have about the same level of accuracy as
D) employ the same methodologies as
E) deal with more comprehensive issues than

12) ________ forecasts are concerned with rates of technological progress, which can result in the birth of exciting
new products, requiring new plants and equipment.
Answer: Technological

13) ________ forecasts address the business cycle by predicting inflation rates, money supplies, housing starts, and
other planning indicators.
Answer: Economic

14) A skeptical manager asks what short-range forecasts can be used for. Give her three possible uses/purposes.
Answer: Any three of: planning purchasing, job scheduling, workforce levels, job assignments, and production
levels.

15) A skeptical manager asks what long-range forecasts can be used for. Give her three possible uses/purposes.
Answer: Any three of: planning for new products, capital expenditures, facility location or expansion, and research
and development.
Section 2: The Strategic Importance of Forecasting

1) What forecasting systems combine the intelligence of multiple supply chain partners?
A) FORE
B) MULTISUP
C) CPFR
D) SUPPLY
E) MSCP

Section 3: Seven Steps in the Forecasting System

1) Forecasts of individual products tend to be more accurate than forecasts of product families.
Answer: FALSE

2) Most forecasting techniques assume that there is some underlying stability in the system.
Answer: TRUE

4-2
3) Which of the following is NOT a step in the forecasting process?
A) Determine the use of the forecast.
B) Eliminate any assumptions.
C) Determine the time horizon of the forecast.
D) Select the forecasting model.
E) Validate and implement the results.

4) Identify the seven steps involved in forecasting.

Answer: 1. Determine the use of the forecast.
2. Select the items to be forecasted.
3. Determine the time horizon of the forecast.
4. Select the forecasting model(s).
5. Gather the data needed to make the forecast.
6. Make the forecast.
7. Validate and implement the results.

5) What are the three realities of forecasting that companies face?

Answer: First, outside factors that we cannot predict or control often impact the forecast. Second, most forecasting
techniques assume that there is some underlying stability in the system. Finally, both product family and aggregated
forecasts are more accurate than individual product forecasts.

Section 4: Forecasting Approaches

1) The sales force composite forecasting method relies on salespersons' estimates of expected sales.
Answer: TRUE

2) A time-series model uses a series of past data points to make the forecast.
Answer: TRUE
3) The quarterly "make meeting" of Lexus dealers is an example of a sales force composite forecast.
Answer: TRUE

4) The two general approaches to forecasting are:

A) qualitative and quantitative.
B) mathematical and statistical.
C) judgmental and qualitative.
D) historical and associative.
E) judgmental and associative.

5) Which of the following uses three types of participants: decision makers, staff personnel, and respondents?
A) jury of executive opinion
B) sales force composite
C) Delphi method
D) associative models
E) time series

6) The forecasting technique that pools the opinions of a group of experts or managers is known as:
A) the expert judgment model.
B) multiple regression.
C) jury of executive opinion.
D) market survey.
E) management coefficients.

7) Which of the following is not a type of qualitative forecasting?

A) jury of executive opinion
B) sales force composite
C) market survey

4-3
D) Delphi method
E) moving average

8) Which of the following techniques uses variables such as price and promotional expenditures, which are related
to product demand, to predict demand?
A) associative models
B) exponential smoothing
C) weighted moving average
D) moving average
E) trend projection

9) ________ forecasts employ one or more mathematical models that rely on historical data and/or associative
variables to forecast demand.
Answer: Quantitative

10) ________ is a forecasting technique based upon salespersons' estimates of expected sales.
Answer: Sales force composite

11) ________ forecasts use a series of past data points to make a forecast.
Answer: Time-series

12) What are the differences between quantitative and qualitative forecasting methods?
Answer: Quantitative methods use mathematical models to analyze historical data. Qualitative methods incorporate
such factors as the decision maker's intuition, emotions, personal experiences, and value systems in determining the
forecast.

13) Identify four quantitative forecasting methods.

Answer: The list includes naive, moving averages, exponential smoothing, trend projection, and linear regression.

14) What is a time-series forecasting model?

Answer: A time-series forecasting model uses a series of past data points to make a forecast.

15) What is the difference between an associative model and a time-series model?
Answer: A time-series model uses only historical values of the quantity of interest to predict future values of that
quantity. The associative model, on the other hand, incorporates the variables or factors that might influence the
quantity being forecast.

16) Name and discuss three qualitative forecasting methods.

Answer: Qualitative forecasting methods include: jury of executive opinion, where high-level managers arrive at a
group estimate of demand; sales force composite, where salespersons' estimates are aggregated; Delphi method,
which uses a group process that allows experts to make forecasts; and market survey, where consumers are queried
about their future purchasing plans.

Section 5: Time-Series Forecasting

1) A naïve forecast for September sales of a product would be equal to the forecast for August.
Answer: FALSE

2) Cycles and random variations are both components of time series.

Answer: TRUE

3) A naïve forecast for September sales of a product would be equal to the sales in August.
Answer: TRUE

4) One advantage of exponential smoothing is the limited amount of record keeping involved.
Answer: TRUE

4-4
5) The larger the number of periods in the simple moving average forecasting method, the greater the method's
responsiveness to changes in demand.
Answer: FALSE

6) Mean squared error and exponential smoothing are two measures of the overall error of a forecasting model.
Answer: FALSE

7) In trend projection, the trend component is the slope of the regression equation.
Answer: TRUE

8) In trend projection, a negative regression slope is mathematically impossible.

Answer: FALSE

9) Seasonal indices adjust raw data for patterns that repeat at regular time intervals.
Answer: TRUE

10) A trend projection equation with a slope of 0.78 means that there is a 0.78 unit rise in Y per period.
Answer: TRUE

11) Demand for individual products can be driven by product life cycles.
Answer: TRUE

12) Which of the following statements about time-series forecasting is true?

A) It is always based on the assumption that future demand will be the same as past demand.
B) It makes extensive use of the data collected in the qualitative approach.
C) It is based on the assumption that the analysis of past demand helps predict future demand.
D) Because it accounts for trends, cycles, and seasonal patterns, it is always more powerful than associative
forecasting.
E) All of the above are true.

13) Time-series data may exhibit which of the following behaviors?

A) trend
B) random variations
C) seasonality
D) cycles
E) They may exhibit all of the above.

14) Gradual upward or downward movement of data over time is called:

A) seasonality.
B) a cycle.
C) a trend.
D) exponential variation.
E) random variation.

15) Which of the following is not present in a time series?

A) seasonality
B) operational variations
C) trend
D) cycles
E) random variations

16) The fundamental difference between cycles and seasonality is the:

A) duration of the repeating patterns.
B) magnitude of the variation.

4-5
C) ability to attribute the pattern to a cause.
D) all of the above
E) none of the above

17) In time series, which of the following cannot be predicted?

A) large increases in demand
B) cycles
C) seasonal fluctuations
D) random variations
E) large decreases in demand

18) What is the forecast for May using a four-month moving average?

Nov. Dec. Jan. Feb. Mar. April

39 36 40 42 48 46

A) 38
B) 42
C) 43
D) 44
E) 47

19) Which time-series model below assumes that demand in the next period will be equal to the most recent period's
demand?
A) naïve approach
B) moving average approach
C) weighted moving average approach
D) exponential smoothing approach
E) trend projection

20) John's House of Pancakes uses a weighted moving average method to forecast pancake sales. It assigns a weight
of 5 to the previous month's demand, 3 to demand two months ago, and 1 to demand three months ago. If sales
amounted to 1000 pancakes in May, 2200 pancakes in June, and 3000 pancakes in July, what should be the forecast
for August?
A) 2400
B) 2511
C) 2067
D) 3767
E) 1622

21) A six-month moving average forecast is generally better than a three-month moving average forecast if demand:
A) is rather stable.
B) has been changing due to recent promotional efforts.
C) follows a downward trend.
D) exceeds one million units per year.
E) follows an upward trend.

22) Increasing the number of periods in a moving average will accomplish greater smoothing, but at the expense of:
A) manager understanding.
B) accuracy.
C) stability.
D) sensitivity to real changes in the data.
E) All of the above are diminished when the number of periods increases.

23) Which of the following statements comparing exponential smoothing to the weighted moving average technique
is TRUE?

4-6
A) Exponential smoothing is more easily used in combination with the Delphi method.
B) More emphasis can be placed on recent values using the weighted moving average.
C) Exponential smoothing is considerably more difficult to implement on a computer.
D) Exponential smoothing typically requires less record keeping of past data.
E) Exponential smoothing allows one to develop forecasts for multiple periods, whereas the weighted moving
average technique does not.

24) Which time-series model uses BOTH past forecasts and past demand data to generate a new forecast?
A) naïve
B) moving average
C) weighted moving average
D) exponential smoothing
E) trend projection

25) Which of the following is NOT a characteristic of exponential smoothing?

A) smoothes random variations in the data
B) uses an easily altered weighting scheme
C) weights each historical value equally
D) has minimal data storage requirements
E) uses the previous period's forecast

26) Which of the following smoothing constants would make an exponential smoothing forecast equivalent to a
naive forecast?
A) 0
B) 1 divided by the number of periods
C) 0.5
D) 1.0
E) cannot be determined
27) Given an actual demand this period of 103, a forecast value for this period of 99, and an alpha of .4, what is the
exponential smoothing forecast for next period?
A) 94.6
B) 97.4
C) 100.6
D) 101.6
E) 103.0

28) A forecast based on the previous forecast plus a percentage of the forecast error is a(n):
A) qualitative forecast.
B) naive forecast.
C) moving average forecast.
D) weighted moving average forecast.
E) exponential smoothing forecast.

29) Given an actual demand this period of 61, a forecast for this period of 58, and an alpha of 0.3, what would the
forecast for the next period be using exponential smoothing?
A) 45.5
B) 57.1
C) 58.9
D) 61.0
E) 65.5

30) Which of the following values of alpha would cause exponential smoothing to respond the SLOWEST to
forecast errors?
A) 0.10
B) 0.2246
C) 0.50

4-7
D) 0.90
E) cannot be determined

31) A forecasting method has produced the following over the past five months. What is the mean absolute
deviation?

Actual Forecast Error |Error|

10 11 -1 1
8 10 -2 2
10 8 2 2
6 6 0 0
9 8 1 1

A) -0.2
B) -1.0
C) 0.0
D) 1.2
E) 8.6

32) The primary purpose of the mean absolute deviation (MAD) in forecasting is to:
A) estimate the trend line.
B) eliminate forecast errors.
C) measure forecast accuracy.
D) seasonally adjust the forecast.
E) remove random variations.

33) Given forecast errors of -1, 4, 8, and -3, what is the mean absolute deviation?
A) 2
B) 3
C) 4
D) 8
E) 16

34) Suppose that the last four months of sales were 8, 10, 15, and 9 units, respectively. Suppose further that the last
four forecasts were 5, 6, 11, and 12 units, respectively. What is the Mean Absolute Deviation (MAD) of these
forecasts?
A) 2
B) -10
C) 3.5
D) 9
E) 10.5

35) A time-series trend equation is 25.3 + 2.1x. What is your forecast for period 7?
A) 23.2
B) 25.3
C) 27.4
D) 40.0
E) 179.2

36) For a given product demand, the time-series trend equation is 53 - 4x. The negative sign on the slope of the
equation:
A) is a mathematical impossibility.
B) is an indication that the forecast is biased, with forecast values lower than actual values.
C) is an indication that product demand is declining.
D) implies that the coefficient of determination will also be negative.
E) implies that the cumulative error will be negative.

4-8
37) Yamaha manufactures which set of products with complementary demands to address seasonal variations?
A) golf clubs and skis
B) swimming suits and winter jackets
C) jet skis and snowmobiles
D) pianos and guitars
E) ice skates and water skis

38) Which of the following is TRUE regarding the two smoothing constants of the Forecast Including Trend (FIT)
model?
A) One constant is positive, while the other is negative.
B) They are called MAD and cumulative error.
C) Alpha is always smaller than beta.
D) One constant smoothes the regression intercept, whereas the other smoothes the regression slope.
E) Their values are determined independently.

39) Demand for a certain product is forecast to be 800 units per month, averaged over all 12 months of the year. The
product follows a seasonal pattern, for which the January monthly index is 1.25. What is the seasonally-adjusted
sales forecast for January?
A) 640 units
B) 798.75 units
C) 801.25 units
D) 1000 units
E) 83.33 units
Answer: D
Diff: 2
Key Term: Seasonal variations
AACSB: Analytic skills
Objective: LO5
Learning Outcome: Describe major approaches to forecasting

40) A seasonal index for a monthly series is about to be calculated on the basis of three years' accumulation of data.
The three previous July values were 110, 150, and 130. The average demand over all months during the three-year
time period was 190 . What is the approximate seasonal index for July?
A) 0.487
B) 0.684
C) 1.462
D) 2.053
E) cannot be calculated with the information given

41) Suppose that the demand in period 1 was 7 units and the demand in period 2 was 9 units. Assume that the
forecast for period 1 was for 5 units. If the firm uses exponential smoothing with an alpha value of .20, what should
be the forecast for period 3? (Round answers to two decimal places.)
A) 9.00
B) 3.72
C) 9.48
D) 5.00
E) 6.12

42) ________ expresses the error as a percent of the actual values.

A) MAD
B) MSE
C) MAPE
D) FIT
E) The smoothing constant

4-9
43) If Brandon Edward were working to develop a forecast using a moving averages approach, but he noticed a
detectable trend in the historical data, he should:
A) use weights to place more emphasis on recent data.
B) use weights to minimize the importance of the trend.
C) change to an associative multiple regression approach.
D) use a simple moving average.
E) change to a qualitative approach.

44) A(n) ________ forecast uses an average of the most recent periods of data to forecast the next period.
Answer moving average (or simple moving average)

45) The smoothing constant is a weighting factor used in ________.

Answer: exponential smoothing

46) A measure of forecast error that does not depend upon the magnitude of the item being forecast is the

Answer: mean absolute percent error (or MAPE)

47) When one constant is used to smooth the forecast average and a second constant is used to smooth the trend, the
forecasting method is ________.
Answer: exponential smoothing with trend adjustment or trend-adjusted smoothing or second-order smoothing or
double smoothing

48) ________ is a time-series forecasting method that fits a trend line to a series of historical data points and then
projects the line into the future for forecasts.
Answer: Trend projection

49) Simple ________ forecasts only work well if we can assume that market demands will stay fairly steady over
time.
Answer: moving average

50) If a barbershop operator noted that Tuesday's business was typically twice as heavy as Wednesday's, and that
Friday's business was typically the busiest of the week, business at the barbershop is subject to
Answer: seasonal variations (or seasonality)

51) Identify the four components of a time series. Which one of these is rarely forecast? Why is this so?
Answer: Trend, seasonality, cycles, and random variation. Since random variations follow no discernible pattern,
they cannot be predicted, and thus are not forecast.

52) Compare seasonal effects and cyclical effects.

Answer: A cycle is longer (typically several years) than a season (typically days, weeks, months, or quarters). A
cycle has variable duration, while a season has fixed duration and regular repetition. Cycles include a wide variety
of factors that cause the economy to go from recession to expansion to recession over a period of years.

53) Distinguish between a weighted moving average model and an exponential smoothing model.
Answer: Exponential smoothing is a weighted moving average model wherein previous values are weighted in a
specific manner--in particular, all previous values are weighted with a set of weights that decline exponentially.
Also, exponential smoothing involves little record keeping of past data.

58) What does it mean to "decompose" a time series?

Answer: To decompose a time series means to break past data down into components of trend, seasonality, cycles,
and random variations, and then to project them forward.

59) What is the key difference between weighted moving average and simple moving average approaches to
forecasting?
Answer: Simple moving averages are useful where there is no identifiable trend in the historical data, i.e., demand

4-10
has been fairly steady over time. If there were an identifiable trend, weighted moving averages would provide a
more accurate forecast because higher weights would be put on the more recent data.

60) Weekly sales of ten-grain bread at the local organic food market are provided in the table below. Based on these
data, forecast week 9 using a five-week moving average.

Week Sales
1 415
2 389
3 420
4 382
5 410
6 432
7 405
8 421
Answer: (382 + 410 + 432 + 405 + 421)/5 = 410.0

61) Given the following data, calculate the three-year moving averages for years 4 through 10.

Year Demand
1 74
2 90
3 59
4 91
5 140
6 98
7 110
8 123
9 99

Answer:
Year Demand 3-Year Moving Ave.
1 74
2 90
3 59
4 91 74.33
5 140 80.00
6 98 96.67
7 110 109.67
8 123 116.00
9 99 110.33
10 110.67

62) What is the forecast for May based on a weighted moving average applied to the following past demand data
and using the weights: 4, 3, 2 (largest weight is for most recent data)?

Nov. Dec. Jan. Feb. Mar. April

37 36 40 42 47 43

Answer: 2 × 42 + 3 × 47 + 4 × 43 = 84 + 141 + 172 = 397; 397/9 = 44.1

63) Weekly sales of copy paper at Cubicle Suppliers are provided in the table below. Compute a three-period

4-11
moving average and a four-period moving average for weeks 5, 6, and 7. Compute the MAD for both forecasting
methods. Which model is more accurate? Forecast week 8 with the more accurate method.

Week Sales (cases)

1 17
2 21
3 27
4 31
5 19
6 17
7 21
Answer:
Week Sales (cases) 3MA |error| 4MA |error|
1 17
2 21
3 27
4 31
5 19 26.3 7.3 24.0 5.0
6 17 25.7 8.7 24.5 7.5
7 21 22.3 1.3 23.5 2.5
8 22.0

The MAD for the 3-week moving average is (7.3 + 8.7 + 1.3) / 3 = 5.77. The MAD for the 4-week moving average
is (5.0 + 7.5 + 2.5) / 3 = 5.00. The four-week moving average is more accurate. The forecast with the 4-moving
average is 22.0.

64) The last four weekly values of sales were 80, 100, 105, and 90 units, respectively. The last four forecasts (for the
same four weeks) were 60, 80, 95, and 75 units, respectively. Calculate the MAD, MSE, and MAPE for these four
weeks.

Sales Forecast Error Error squared Pct. error

80 60 20 400 .25
100 80 20 400 .20
105 95 10 100 .095
90 75 15 225 .167

Answer: MAD = 65/4 = 16.25; MSE = 1125/4 = 281.25; MAPE = 0.712/4 = .178 or 17.8%
65) A management analyst is using exponential smoothing to predict merchandise returns at an upscale branch of a
department store chain. Given an actual number of returns of 154 items in the most recent period completed, a
forecast of 172 items for that period, and a smoothing constant of 0.3, what is the forecast for the next period? How
would the forecast be changed if the smoothing constant were 0.6? Explain the difference in terms of alpha and
responsiveness.
Answer: 166.6; 161.2 The larger the smoothing constant in an exponentially smoothed forecast, the more
responsive the forecast.

67) Favors Distribution Company purchases small imported trinkets in bulk, packages them, and sells them to retail
stores. The managers are conducting an inventory control study of all their items. The following data are for one
such item, which is not seasonal.

a. Use a trend projection to estimate the relationship between time and sales (state the equation).
b. Calculate forecasts for the first four months of the next year.

1 2 3 4 5 6 7 8 9 10 11 12
Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Sales 51 55 54 57 50 68 66 59 67 69 75 73

4-12
Answer: The trend projection equation is y-hat = 48.32 + 2.105x. The next four months are forecast to be 75.68,
77.79, 79.89, and 82.00

68) Use exponential smoothing with trend adjustment to forecast deliveries for period 10. Let alpha = 0.4, beta =
0.2, and let the initial trend value be 4 and the initial forecast be 200.

Actual
Period Demand
1 200
2 212
3 214
4 222
5 236
6 221
7 240
8 244
9 250
10 266

Answer:
Actual Forecast Trend FIT
1 200 200.00 4.00
2 212 202.40 3.68 206.08
3 214 208.45 4.15 212.60 70) Demand for a certain product is forecast to
4 222 213.16 4.27 217.43 be 8,000 units per month, averaged over all 12
5 236 219.26 4.63 223.89 months of the year. The product follows a
6 221 228.73 5.60 234.33 seasonal pattern, for which the January
7 240 229.00 4.53 233.53 monthly index is 1.25. What is the seasonally-
8 244 236.12 5.05 241.17 adjusted sales forecast for January?
Answer: 8,000 × 1.25 = 10,000
9 250 242.30 5.28 247.58
10 266 248.55 5.47 254.02
71) A seasonal index for a monthly series is
about to be calculated on the basis of three years' accumulation of data. The three previous July values were 110,
135, and 130. The average over all months is 160. What is the approximate seasonal index for July?
Answer: (110 + 135 + 130)/3 = 125; 125/160 = 0.781

74) The last seven weeks of demand at a new car dealer are shown below. Use a three-period weighted-moving
average forecast to determine a forecast for the 8th week using weights of 3, 2, and 1 (where the most recent week
receives the highest weight). (Round all forecasts to the nearest whole unit.) Calculate the MAD for this forecast
(covering all weeks in which error comparisons can be made). What does the MAD indicate?

Week Sales
1 25
2 30
3 27
4 31
5 27
6 29
7 30

MAD = 8/4 = 2

75) A small family-owned restaurant uses a seven-day moving average model to determine manpower requirements.

4-13
These forecasts need to be seasonalized because each day of the week has its own demand pattern. The seasonal
indices for each day of the week are: Monday 0.445; Tuesday 0.791; Wednesday 0.927; Thursday 1.033; Friday
1.422; Saturday 1.478; and Sunday 0.903. Average daily demand based on the most recent moving average is 194
patrons. What is the seasonalized forecast for each day of next week?
Answer: The average value multiplied by each day's seasonal index. Monday: 194 × .445 = 86; Tuesday: 194 × .791
= 153; Wednesday: 194 × .927 = 180; Thursday: 194 × 1.033 = 200; Friday: 194 × 1.422 = 276; Saturday: 194 ×
1.478 = 287; and Sunday: 194 × .903 = 175.

76) The department manager using a combination of methods has forecast sales of toasters at a local department
store. Calculate the MAD for the manager's forecast. Compare the manager's forecast against a naive forecast
covering the same time period. Which is better?

Manager's
Month Unit Sales Forecast
January 52
February 61
March 73
April 79
May 66
June 51
July 47 50
August 44 55
September 30 52
October 55 42
November 74 60
December 125 75

The manager's forecast has a MAD of 18.83, while the naive is 19.33. Therefore, the manager's forecast is slightly
better than the naive.

Section 6 Associative Forecasting Methods: Regression and Correlation Analysis

1) Linear-regression analysis is a straight-line mathematical model to describe the functional relationships between
independent and dependent variables.
Answer: TRUE

2) The larger the standard error of the estimate, the more accurate the forecasting model.
Answer: FALSE

3) In a regression equation where y-hat is demand and x is advertising, a coefficient of determination (R2) of .70
means that 70% of the variance in advertising is explained by demand.
Answer: FALSE

4) Regression lines graphically depict "cause-and-effect" relationships.

Answer: FALSE
Diff: 3
Key Term: Linear-regression analysis
Objective: LO6
Learning Outcome: Describe major approaches to forecasting

5) A fundamental distinction between trend projection and linear regression is that:

A) trend projection uses least squares while linear regression does not.
B) only linear regression can have a negative slope.
C) in trend projection the independent variable is time; in linear regression the independent variable need not be
time, but can be any variable with explanatory power.

4-14
D) trend projection can be a function of several variables, while linear regression can only be a function of one
variable.
E) trend projection uses two smoothing constants, not just one.

6) The degree or strength of a relationship between two variables is shown by the:

A) alpha.
B) mean.
C) mean absolute deviation.
D) coefficient of correlation.
E) cumulative error.

7) If two variables were perfectly correlated, what would the coefficient of correlation r equal?
A) 0
B) -1
C) 1
D) B or C
E) none of the above

8) Linear regression is known as a(n) ________ model because it incorporates variables or factors that might
influence the quantity being forecast.
Answer: associative forecasting

9) The ________ measures the strength of the relationship between two variables.
Answer: coefficient of correlation

10) Distinguish a dependent variable from an independent variable.

Answer: The dependent variable is the factor or behavior that we are trying to explain or predict. The independent
variables are believed to impact the dependent variable and thus affect its value.

12) A firm has modeled its experience with industrial accidents and found that the number of accidents per year (y-
hat) is related to the number of employees (x) by the regression equation:
y-hat = 3.3 + 0.049x. The r-squared value is 0.68. The regression is based on 20 annual observations. The firm
intends to employ 480 workers next year. How many accidents do you project? How much confidence do you have
in that forecast?
Answer: y-hat = 3.3 + 0.049(480) = 3.3 + 23.52 = 26.82 accidents. This is not a time series, so next year = year 21
is of no relevance. Confidence comes from the coefficient of determination; the model explains 68% of the variation
in number of accidents, which seems respectable.

Section 7: Monitoring and Controlling Forecasts

1) If a forecast is consistently greater than (or less than) actual values, the forecast is said to be biased.
Answer: TRUE

2) Focus forecasting tries a variety of computer models and selects the best one for a particular application.
Answer: TRUE
3) The last four weekly values of sales were 80, 100, 105, and 90 units. The last four forecasts were 60, 80, 95, and
75 units. These forecasts illustrate:
A) qualitative methods.
B) adaptive smoothing.
C) slope.
D) bias.
E) trend projection.

4) The tracking signal is the:

A) standard error of the estimate.
B) absolute deviation of the last period's forecast.

4-15
C) MAD.
D) ratio of cumulative error / MAD.
E) MAPE.

5) Computer monitoring of tracking signals and self-adjustment if a signal passes a preset limit is characteristic of:
A) exponential smoothing including trend.
B) adaptive smoothing.
C) trend projection.
D) focus forecasting.
E) multiple regression analysis.

6) ________ forecasting tries a variety of computer models and selects the best one for a particular application.
Answer: Focus

7) An approach to exponential smoothing in which the smoothing constant is automatically changed to keep errors to
a minimum is called ________.
Answer: adaptive smoothing

8) What is a tracking signal? Explain the connection between adaptive smoothing and tracking signals.
Answer: A tracking signal is a measure of how well the forecast actually predicts. The larger the absolute tracking
signal, the worse the forecast is performing. Adaptive smoothing sets limits for the tracking signal and makes
changes to its forecasting models when the tracking signal goes beyond those limits.

10) Jim's department at a local department store has tracked the sales of a product over the last ten weeks. Forecast
demand using exponential smoothing with an alpha of 0.4, and an initial forecast of 28.0 for period 1. Calculate the
MAD. Calculate the tracking signal. What do you recommend?

Period Demand
1 24
2 23
3 26
4 36
5 26
6 30
7 32
8 26
9 25
10 28

Answer:
Period Demand Forecast Error Absolute
1 24 28.00
2 23 26.40 -3.40 3.40
3 26 25.04 0.96 0.96
4 36 25.42 10.58 10.58
5 26 29.65 -3.65 3.65
6 30 28.19 1.81 1.81
7 32 28.92 3.08 3.08
8 26 30.15 -4.15 4.15
9 25 28.49 -3.49 3.49
10 28 27.09 0.91 0.91
Total 2.64 32.03
Average 0.29 3.56
Bias MAD

4-16
The tracking signal RSFE/MAD = 2.64/3.56 = .742 is low; therefore, keep using the forecasting method.
Section 8: Forecasting in the Service Sector

1) Many services maintain records of sales noting:

A) the day of the week.
B) unusual events.
C) the weather.
D) holiday impacts.
E) all of the above.

2) Taco Bell's unique employee scheduling practices are partly the result of using:
A) point-of-sale computers to track food sales in 15 minute intervals.
B) focus forecasting.
C) a six-week moving average forecasting technique.
D) multiple regression.
E) A and C are both correct.
Answer: E
Diff: 2
Learning Outcome: Describe major approaches to forecasting

4-17