Deck 5: Forecasting

Question 1

Business forecasting is an estimation of future events.

Answer 1

False

Question 2

Forecasting offers a logical basis for making business decisions.

Answer 2

True

Question 3

Business forecasting is a probabilistic statement about the state of the future.

Answer 3

True

Question 4

Forecasting should only be used for operations.

Question 5

The forecasting method used does not depend on the time horizon over which you are forecasting.

Question 6

It is always necessary to measure the accuracy of and make corrections to a forecast.

Question 7

Items with lower volatility can be less accurately forecasted.

Question 8

The further out the time horizon, the more accurate the forecast will be.

Question 9

Quantitative forecasting tends to be more accurate than qualitative forecasting.

Question 10

Forecasts of individual products tend to be more accurate than aggregated forecasts.

Question 11

Most forecasting methods assume that there is some underlying stability in the system.

Question 12

Market surveys are the most frequently used quantitative method of forecasting.

Question 13

Time series techniques are most commonly used for business forecasting.

Question 14

The sales force estimate forecasting method relies on salespersons' estimates of expected sales.

Question 15

The meeting of operations, purchasing, accounting, finance, and selected key suppliers and customers is an example of the jury of executive opinion forecasting.

Question 16

A consumer survey is one of the most reliable forecasting methods.

Question 17

A naïve forecast for June sales of a product would be equal to the sales in May.

Question 18

Moving averages and exponential smoothing are examples of time series analysis.

Question 19

Time series decomposition breaks down the data into seasonal patterns only.

Question 20

An advantage of single exponential smoothing is the minimal amount of data required.

Question 21

Methods such as moving averages and single exponential smoothing are ideal for data with a strong trend pattern.

Question 22

Forecast error is the forecasted value minus the actual value.

Question 23

A negative forecast error is better than a positive forecast error.

Question 24

Only the mean absolute percentage error can be used to compare errors across different data sets.

Question 25

For the weighted moving average method, the weights do not have to sum up to one.

Question 26

All data sets used in time series decomposition have a random component.

Question 27

In trend projection, the trend component is the slope of the regression line.

Question 28

In trend analysis, a negative regression slope is mathematically impossible.

Question 29

Patterns in the data that occur every several years are called outliers.

Question 30

Leading indicators change after changes in a business cycle.

Question 31

Lagging indicators change prior to changes in a business cycle.

Question 32

A forecast is said to be biased if it is consistently greater than actual values.

Question 33

The larger the R² value obtained in the linear regression output, the more accurate your forecasts will be.

Question 34

Combining forecast methods by some means always results in lesser accuracy.

Question 35

Technological forecasts tend to be qualitative in nature.

Question 36

Forecasting ‛black swans' are not difficult to do because they can easily be spotted.

Question 37

Forecasting sales is the same as forecasting demand.

Question 38

Data mining using computing power and sophisticated analytical methods enables the discovery of otherwise hidden relationships among variables of interest.

Question 39

Cloud computing is a recent trend in forecasting demand.

Question 40

Wintel Electronics is a new startup and wants to forecast sales of their newest device for the next 3 months using the naïve extrapolation method. What information would they need to make the forecast?

A) last month's actual sales
B) the last 3 months' forecasted sales and 3 months' actual sales
C) the last 3 months' sales
D) last month's forecasted sales

Question 41

Forecasts

A) become more accurate with longer time horizons
B) are more accurate for individual items than for groups of items
C) are seldom perfect
D) all of the above

Question 42

One use of short-term forecasts is to determine

A) the price of gold over the next five years
B) the sales of the newest iWear glasses
C) worker assignments
D) facility location

Question 43

There are three categories of forecasts classified by time horizons

A) short-term, medium-term, and long-term
B) exponential smoothing, regression, and time series
C) strategic, analytical, and operational
D) finance, marketing, and operations

Question 44

A forecast with a time frame of 3 months to 2 years is typically a

A) strategic forecast
B) long-term forecast
C) short-term forecast
D) medium-term forecast

Question 45

Firms doing facility location or expansion, and R&D typically utilize

A) medium-term forecasting
B) long-term forecasting
C) short-term forecasting
D) all of the above

Question 46

Which of the following forecasting methods uses top experts in the field?

A) executive opinions
B) the Delphi method
C) sales force composites
D) quantitative analysts

Question 47

The forecasting method that uses the opinions of a group of experts or managers is known as

A) the jury of executive opinion model
B) the management model
C) the Delphi method
D) the expert judgment model

Question 48

Which of the following is not a type of quantitative forecasting?

A) single linear regression
B) moving average method
C) the Delphi method
D) time series analysis

Question 49

All of the following are types of qualitative forecasting except

A) the Delphi method
B) jury of executive opinions
C) sales force estimation
D) moving average method

Question 50

Time series data have all of the following properties except

A) circles
B) random variations
C) seasonality
D) patterns

Question 51

Which of the following statements about time series analysis is true?

A) It is popular with businesses.
B) It is based on the assumption that future demand will be the same as past demand.
C) It only uses the most recent data collected.
D) It is based on moving averages only.

Question 52

What is the approximate forecast for July using a 4-month moving average?
$$\begin{array} { c c c c c c } \text { Jan. } & \text { Feb. } & \text { Mar. } & \text { April } & \text { May } & \text { June } \\49 & 46 & 50 & 52 & 58 & 56\end{array}$$

A) 54
B) 48
C) 57
D) 52

Question 53

Which forecasting technique below assumes that demand in the next period will be equal to the most recent period's demand?

A) weighted moving average
B) moving average
C) exponential smoothing
D) naïve method

Question 54

Kim's Nail Salon uses a weighted moving average method to forecast demand. She assigns a weight of 5 to the previous month's demand, 3 to demand two months ago, and 2 to demand three months ago. If demand was 800 customers in April, 900 customers in May, 1,200 customers in June, and 2,000 customers in July, what should her forecast for August be?

A) 1,540
B) 910
C) 1,400
D) 1,210

Question 55

For the weighted moving average forecasting method, the weights are determined

A) arbitrarily
B) by operations managers
C) by experience
D) all of the above

Question 56

Which of the following statements comparing the weighted moving average method and the exponential smoothing method is true?

A) Exponential smoothing typically uses less past data.
B) Exponential smoothing is more accurate.
C) For the weighted moving average method, you can use a single weight.
D) More emphasis can be placed on recent values using the weighted moving average.

Question 57

For exponential smoothing, you need

A) past forecast only
B) past demand only
C) past forecast and past demand
D) none of the above

Question 58

Which of the following values of the exponential smoothing constant would emphasize the most recent forecast?

A) 1
B) 0
C) 2.0
D) cannot be determined from the information provided

Question 59

Given an actual demand of 100, a previous forecast value of 90, and an alpha of 0.2, the exponential smoothing forecast for the next period would be

A) 92.6
B) 92.0
C) 118.0
D) 98.0

Question 60

Data from the previous 5 months of a forecasting method is shown below. What is the mean absolute deviation? $$\begin{array} { | c | c | c | c | } \hline \text { Actual } & \text { Forecast } & \text { Error } & \mid \text { Error } \mid \\\hline 12 & 13 & - 1 & 1 \\\hline 10 & 12 & - 2 & 2 \\\hline 12 & 10 & 2 & 2 \\\hline 8 & 8 & 0 & 0 \\\hline 11 & 10 & 1 & 1 \\\hline\end{array}$$

A) −1.2
B) 0.0
C) 6.0
D) 1.2

Question 61

A time series trend equation is 15.2 + 4.1X. What is your forecast for period 7?

A) cannot be determined without knowing the value of X
B) 43.9
C) 15.2
D) 26.3

Question 62

Which of the following error measurement can be used to compare forecasts across different data sets?

A) MAD
B) MSE
C) MAPE
D) none of the above

Question 63

Demand for Honda hybrids nationwide is forecast to be 600 cars per month, averaged over all 12 months of the year. The product follows a seasonal pattern, for which the July monthly index is 1.25. What is the seasonally adjusted sales forecast for July?

A) 750 cars
B) 480 cars
C) 5,760 cars
D) 9,000 cars

Question 64

A data set has been observed to demonstrate seasonality and the three previous yearly January values were 130, 200, and 120. The average over all months is 150. The approximate seasonal index for January is therefore

A) 0.867
B) 1.667
C) 0.800
D) 1.000

Question 65

The larger the value of R² in linear regression implies the

A) better the explanatory power of the model
B) greater the F statistic
C) smaller the p-value
D) none of the above

Question 66

The last four monthly sales data were 100, 120, 125, and 135 units. The last four forecasts were 80, 100, 105, and 115 units. These forecasts illustrate

A) exponential smoothing
B) a trend projection method
C) bias
D) qualitative methods

Question 67

Which of the following utilizes short-term forecasting (as opposed to long-term or medium-term forecasting)?

A) capital expenditures
B) production levels
C) job scheduling
D) location decisions

Question 68

Forecasts can go wrong because of

A) random events
B) white swans
C) calculation errors
D) incorrect data

Question 69

Black swans are

A) found only in Australia
B) found in Australia and Japan
C) birds with fat tails
D) very low probability, previously unobserved events

Question 70

Recent trends in forecasting include

A) data mining
B) crowd forecasting
C) artificial Intelligence
D) all of the above

Question 71

What is the qualitative forecasting model and when is it appropriate?

Question 72

Identify briefly and describe the two general forecasting approaches.

Question 73

Identify the three forecasting time horizons. State an approximate duration of each.

Question 74

Briefly describe the steps that are used to develop a forecasting system.

Question 75

A skeptical manager asks what medium-range forecasts can be used for. Give him three possible uses.

Question 76

Explain why such forecasting devices as moving averages, weighted averages, and exponential smoothing are not well suited for data series with trends.

Question 77

What is the basic difference between a weighted moving average and an exponential smoothing?

Question 78

What three models are used to determine the accuracy of any given forecasting method? How would you determine whether time series regression or exponential smoothing is better in a specific application?

Question 79

Briefly describe the Delphi method. How would it be used for an employer you have worked for?

Question 80

What is the primary difference between a time series model and a causal model?