Chapter 03 Forecasting

1. Which of the following is a potential shortcoming of using sales force opinions in demand forecasting?

A. Members of the sales force often have substantial histories of working with and understanding their customers.
B. Members of the sales force often are well aware of customers' future plans.
C. Members of the sales force have direct contact with consumers.
D. Members of the sales force can have difficulty distinguishing between what customers would like to do and what they
actually will do.
E. Customers often are quite open with members of the sales force with regard to future plans.

Customers themselves may be unclear regarding what they'd like to do versus what they'll actually do.

2. Suppose a four-period weighted average is being used to forecast demand. Weights for the periods are as follows: w t-4 = 0.1,
wt-3 = 0.2, wt-2 = 0.3 and wt-1 = 0.4. Demand observed in the previous four periods was as follows: At-4 = 380, At-3 = 410,
At-2 = 390, At-1 = 400. What will be the demand forecast for period t?

A. 402
B. 397
C. 399
D. 393
E. 403

The forecast will be (.1 * 380) + (.2 * 410) + (.3 * 390) + (.4 * 400) = 397.

3. Suppose a three-period weighted average is being used to forecast demand. Weights for the periods are as follows: w t-3 =
0.2, wt-2 = 0.3 and wt-1 = 0.5. Demand observed in the previous three periods was as follows: At-3 = 2,200, At-2 = 1,950, At-1
= 2,050. What will be the demand forecast for period t?

A. 2,000
B. 2,095
C. 1,980
D. 2,050
E. 1,875

The forecast for will be (.2 * 2,200) + (.3 * 1,950) + (.5 * 2,050) = 2,050.

4. When choosing a forecasting technique, a critical trade-off that must be considered is that between:

A. time series and associative.

B. seasonality and cyclicality.
C. length and duration.
D. simplicity and complexity.
E. cost and accuracy.

The trade-off between cost and accuracy is the critical consideration when choosing a forecasting technique.

3-1
5. The more novel a new product or service design is, the more forecasters have to rely on:

A. subjective estimates.
B. seasonality.
C. cyclicality.
D. historical data.
E. smoothed variation.

New products and services lack historical data, so forecasts for them must be based on subjective estimates.

6. Which of the following is/are a primary input into capacity, sales, and production planning?

A. product design
B. market share
C. ethics
D. globalization
E. demand forecasts

Demand forecasts are direct inputs into capacity, sales, and production plans.

7. Which of the following features would not generally be considered common to all forecasts?

A. Assumption of a stable underlying causal system.

B. Actual results will differ somewhat from predicted values.
C. Historical data is available on which to base the forecast.
D. Forecasts for groups of items tend to be more accurate than forecasts for individual items.
E. Accuracy decreases as the time horizon increases.

In some forecasting situations historical data are not available.

8. Minimizing the sum of the squared deviations around the line is called:

A. mean squared error technique.

B. mean absolute deviation.
C. double smoothing.
D. least squares estimation.
E. predictor regression.

Least squares estimations minimize the sum of squared deviations around the estimated regression function.

3-2
9. The two general approaches to forecasting are:

A. mathematical and statistical.

B. qualitative and quantitative.
C. judgmental and qualitative.
D. historical and associative.
E. precise and approximation.

Forecast approaches are either quantitative or qualitative.

10. Which of the following is not a type of judgmental forecasting?

A. executive opinions
B. sales force opinions
C. consumer surveys
D. the Delphi method
E. time series analysis

Time series analysis is a quantitative approach.

11. Accuracy in forecasting can be measured by:

A. MSE.
B. MRP.
C. MPS.
D. MTM.
E. MTE.

MSE is mean squared error.

12. The forecasting method which uses anonymous questionnaires to achieve a consensus forecast is:

A. sales force opinions.

B. consumer surveys.
C. the Delphi method.
D. time series analysis.
E. executive opinions.

Anonymity is important in Delphi efforts.

13. Detecting nonrandomness in errors can be done using:

A. MSEs.
B. MAPs.
C. control charts.
D. correlation coefficients.
E. strategies.

3-3
 Control charts graphically depict the statistical behavior of forecast errors.

14. Gradual, long-term movement in time series data is called:

A. seasonal variation.
B. cycles.
C. irregular variation.
D. trend.
E. random variation.

Trends move the time series in a long-term direction.

15. The primary difference between seasonality and cycles is:

A. the duration of the repeating patterns.

B. the magnitude of the variation.
C. the ability to attribute the pattern to a cause.
D. the direction of the movement.
E. there are only four seasons but 30 cycles.

Seasons happen within time periods; cycles happen across multiple time periods.

16. Averaging techniques are useful for:

A. distinguishing between random and nonrandom variations.

B. smoothing out fluctuations in time series.
C. eliminating historical data.
D. providing accuracy in forecasts.
E. average people.

Smoothing helps forecasters see past random error.

17. Using the latest observation in a sequence of data to forecast the next period is:

A. a moving average forecast.

B. a naive forecast.
C. an exponentially smoothed forecast.
D. an associative forecast.
E. regression analysis.

Only one piece of information is needed for a naive forecast.

18. For the data given below, what would the naive forecast be for period 5?

Period Value
1 58

3-4
 2 59
3 60
4 61

A. 58
B. 62
C. 59.5
D. 61
E. cannot tell from the data given

Period 5's forecast would be period 4's demand.

19. Moving average forecasting techniques do the following:

A. Immediately reflect changing patterns in the data.

B. Lead changes in the data.
C. Smooth variations in the data.
D. Operate independently of recent data.
E. Assist when organizations are relocating.

Variation is smoothed out in moving average forecasts.

20. A forecast based on the previous forecast plus a percentage of the forecast error is:

A. a naive forecast.
B. a simple moving average forecast.
C. a centered moving average forecast.
D. an exponentially smoothed forecast.
E. an associative forecast.

Exponential smoothing uses the previous forecast error to shape the next forecast.

21. Which is not a characteristic of exponential smoothing?

A. smoothes random variations in the data

B. weights each historical value equally
C. has an easily altered weighting scheme
D. has minimal data storage requirements
E. smoothes real variations in the data

The most recent period of demand is given the most weight in exponential smoothing.

3-5
22. Which of the following smoothing constants would make an exponential smoothing forecast equivalent to a naive forecast?

A. 0
B. .01
C. .1
D. .5
E. 1.0

An alpha of 1.0 leads to a naive forecast.

23. Which of the following possible values of alpha would cause exponential smoothing to respond the most quickly to forecast
errors?

A. 0
B. .01
C. .05
D. .10
E. .15

Larger values for alpha correspond with greater responsiveness.

24. A manager uses the following equation to predict monthly receipts: Y t = 40,000 + 150t. What is the forecast for July if t = 0
in April of this year?

A. 40,450
B. 40,600
C. 42,100
D. 42,250
E. 42,400

July would be period 3, so the forecast would be 40,000 + 150(3).

25. In the additive model for seasonality, seasonality is expressed as a ______________ adjustment to the average; in the
multiplicative model, seasonality is expressed as a __________ adjustment to the average.

A. quantity; percentage
B. percentage; quantity
C. quantity; quantity
D. percentage; percentage
E. qualitative; quantitative

The additive model simply adds a seasonal adjustment to the deseasonalized forecast. The multiplicative model adjusts the
deseasonalized forecast by multiplying it by a season relative or index.

26. A persistent tendency for forecasts to be greater than or less than the actual values is called:

A. bias.

3-6
 B. tracking.
C. control charting.
D. positive correlation.
E. linear regression.

Bias is a tendency for a forecast to be above (or below) the actual value.

27. The primary method for associative forecasting is:

A. sensitivity analysis.
B. regression analysis.
C. simple moving averages.
D. centered moving averages.
E. exponential smoothing.

Regression analysis is an associative forecasting technique for fitting a line to a set of points.
28. The mean absolute deviation is used to:

A. estimate the trend line.

B. eliminate forecast errors.
C. measure forecast accuracy.
D. seasonally adjust the forecast.
E. compute periodic forecast errors.

MAD is one way of evaluating forecast performance.

29. Given forecast errors of 4, 8, and -3, what is the mean absolute deviation?

A. 4
B. 3
C. 5
D. 6
E. 12

Convert each error into an absolute value and then average.

30. Which of the following is used for constructing a control chart?

A. mean absolute deviation

B. mean squared error
C. tracking signal
D. bias
E. MAPE

The mean squared error leads to an estimate for the sample forecast standard deviation.

3-7
31. The two most important factors in choosing a forecasting technique are:

A. cost and time horizon.

B. accuracy and time horizon.
C. cost and accuracy.
D. quantity and quality.
E. objective and subjective components.

More accurate forecasts cost more but may not be worth the additional cost.

32. Given the following historical data, what is the simple three-period moving average forecast for period 6?

Period Value
1 73
2 68
3 65
4 72
5 67

A. 67
B. 115
C. 69
D. 68
E. 68.67

Average demand from periods 3 through 5.

33. Given the following historical data and weights of .5, .3, and .2, what is the three-period moving average forecast for period
5?

Period Value Period Value

1 138 3 148
2 142 4 144

A. 144.20
B. 144.80
C. 144.67
D. 143.00
E. 144.00

Multiply period 4 (144) by .5, period 3 (148) by .3, and period 2 (142) by .2, then sum these products.

3-8
34. Use of simple linear regression analysis assumes that:

A. variations around the line are nonrandom.

B. deviations around the line are normally distributed.
C. predictions can easily be made beyond the range of observed values of the predictor variable.
D. all possible predictor variables are included in the model.
E. the variance of error terms (deviations) varies directly with the predictor variable.
That deviations conform to the normal distribution is a very important assumption underpinning simple linear regression.

35. Given forecast errors of -5, -10, and +15, the MAD is:

A. 0.
B. 10.
C. 30.
D. 175.
E. 225.

Convert these errors into absolute value, then average.

36. The president of State University wants to forecast student enrollments for this academic year based on the following
historical data:

Year Enrollments
5 Years ago 15,000
4 Years ago 16,000
3 Years ago 18,000
2 Years ago 20,000
Last Year 21,000

What is the forecast for this year using the naive approach?

A. 18,750
B. 19,500
C. 21,000
D. 22,000
E. 22,800

There is a clear trend with continuously increasing enrollments, so this year's forecast would be last year's enrollment plus
the difference between last year’s enrollment and the enrollment two years ago. 21000 + (21000 - 20000) = 22000

37. The president of State University wants to forecast student enrollments for this academic year based on the following
historical data:

Year Enrollments
5 Years ago 15,000
4 Years ago 16,000
3 Years ago 18,000
2 Years ago 20,000
Last Year 21,000
3-9
 What is the forecast for this year using a four-year simple moving average?

A. 18,750
B. 19,500
C. 21,000
D. 22,650
E. 22,800

Average enrollment from the last four years.

(21000 + 20000 + 18000 + 16000) / 4 = 18750

38. The business analyst for Video Sales, Inc. wants to forecast this year's demand for DVD decoders based on the following
historical data:

Year Demand
5 Years ago 900
4 Years ago 700
3 Years ago 600
2 Years ago 500
Last Year 300

What is the forecast for this year using the naive approach?

A. 100
B. 200
C. 300
D. 500
E. 600

There is a clear trend with continuously decreasing demand, so this year's forecast is last year's demand plus the difference
between last year’s demand and the demand two years ago. 300 + (300 - 500) = 100

39. The business analyst for Video Sales, Inc. wants to forecast this year's demand for DVD decoders based on the following
historical data:

Year Enrollments
5 Years ago 900
4 Years ago 700
3 Years ago 600
2 Years ago 500
Last Year 300

What is the forecast for this year using a three-year weighted moving average with weights of .5, .3, and .2?

A. 163
3-10
 B. 180
C. 300
D. 420
E. 510

Multiply last year (300) by .5, 2 years ago (500) by .3, and 3 years ago (600) by .2, then sum these products. 0.5*300 +
0.3*500 + 0.2*600 = 420

40. Professor Very Busy needs to allocate time next week to include time for office hours. He needs to forecast the number of
students who will seek appointments. He has gathered the following data:

Week #Students
6 Weeks ago 83
5 Weeks ago 110
4 Weeks ago 95
3 Weeks ago 80
2 Weeks ago 65
Last Year 50

What is this week's forecast using the naive approach?

A. 35
B. 50
C. 52
D. 65
E. 78

Other than the original data point 6 weeks ago, there is a clear trend over the last 5 weeks with continuously decreasing
demand, so this week's forecast is last week's demand plus the difference between last week’s demand and the demand two
weeks ago. 50 + (50 - 65) = 35

41. Professor Very Busy needs to allocate time next week to include time for office hours. He needs to forecast the number of
students who will seek appointments. He has gathered the following data:

Week #Students
6 Weeks ago 83
5 Weeks ago 110
4 Weeks ago 95
3 Weeks ago 80
2 Weeks ago 65
Last Week 50

What is this week's forecast using a three-week simple moving average?

A. 49
B. 50
C. 52
3-11
 D. 65
E. 78

Average the three most recent weeks of demand. (50 + 65 + 80) / 3 = 65

42. A concert promoter is forecasting this year's attendance for one of his concerts based on the following historical data:

Year Attendance
Four Years ago 10,000
Three Years ago 12,000
Two Years ago 18,000
Last Year 20,000

What is this year's forecast using the naive approach?

A. 22,000
B. 20,000
C. 18,000
D. 15,000
E. 12,000

There is a clear trend with continuously increasing attendance, so this year's forecast would be last year's attendance plus the
difference between last year’s attendance and the attendance two years ago. 20000 + (20000 – 18000) = 22000

43. A concert promoter is forecasting this year's attendance for one of his concerts based on the following historical data:

Year Attendance
Four Years ago 10,000
Three Years ago 12,000
Two Years ago 18,000
Last Year 20,000

What is this year's forecast using a two-year weighted moving average with weights of .7 and .3?

A. 19,400
B. 18,600
C. 19,000
D. 11,400
E. 10,600

Multiply last year (20000) by .7 and 2 years ago (18000) by .3, then sum these products. 0.7*20000 + 0.3*18000 = 19400
44. The dean of a school of business is forecasting total student enrollment for this year's summer session classes based on the
following historical data:

Year Enrollment
Four years ago 2000
Three years ago 2200
Two years ago 2800
Last year 3000

3-12
 What is this year's forecast using the naive approach?

A. 2,000
B. 2,200
C. 2,800
D. 3,000
E. 3,200

There is a clear trend with continuously increasing enrollments, so this year's forecast would be last year's enrollment plus
the difference between last year’s enrollment and the enrollment two years ago. 3000 + (3000 – 2800) = 3200

45. The dean of a school of business is forecasting total student enrollment for this year's summer session classes based on the
following historical data:

Year Enrollment
Four years ago 2000
Three years ago 2200
Two years ago 2800
Last year 3000

What is this year's forecast using a three-year simple moving average?

A. 2,667
B. 2,600
C. 2,500
D. 2,400
E. 2,333

Average the most recent periods of enrollment. (3000 + 2800 + 2200) / 3 = 2667
46. The owner of Darkest Tans Unlimited in a local mall is forecasting this month's (October's) demand for the one new tanning
booth based on the following historical data:

Month Number of Visits

April 100
May 140
June 110
July 150
August 120
September 160

What is this month's forecast using the naive approach?

A. 100
B. 160
C. 130
D. 140
E. 120

There is no obvious trend or seasonality, so this month's forecast is last month's demand.
3-13
47. The owner of Darkest Tans Unlimited in a local mall is forecasting this month's (October's) demand for the one new tanning
booth based on the following historical data:

Month Number of Visits

April 100
May 140
June 110
July 150
August 120
September 160

What is this month's forecast using a four-month weighted moving average with weights of .4, .3, .2, and .1?

A. 120
B. 129
C. 141
D. 135
E. 140

Multiply September (160) by .4, August (120) by .3, July (150) by .2, and June (110) by .1, then sum these products. 0.4*160
+ 0.3*120 + 0.2*150 + 0.1*110 = 141

48. Which of the following helps improve supply chain forecasting performance?

A. contracts that require supply chain members to formulate long-term forecasts

B. penalties for supply chain members that adjust forecasts
C. sharing forecasts or demand data across the supply chain
D. increasing lead times for critical supply chain members
E. increasing the number of suppliers at critical junctures in the supply chain

Sharing forecasts and/or demand data is a means of ensuring that the supply chain's overall forecast is as accurate as it can
be.

3-14