Deck 14: Forecasting Demand for Services

Question 1

The key feature of the historical analogy method is that it assumes some future event is related to the occurrence of an earlier event.

Answer 1

False

Question 2

Exponential smoothing prevents overreaction to extremes in the actual observed values.

Answer 2

True

Question 3

Costs for preparing time-series forecasts generally are lower than for other models.

Answer 3

True

Question 4

The greatest advantage of subjective techniques like Delphi and cross-impact analyses, is that the methods are fairly standardized and do not require much expertise in actual use.

Question 5

Because of the nontangible nature of a service, forecasting does not play as important a role in service operations as in manufacturing operations.

Question 6

Both measures of forecast accuracy, MAD and MSD, give equal weight to all forecasting errors.

Question 7

The trade-off to be made with respect to accuracy is between the costs of inaccurate forecasts and the costs of increasing the accuracy of forecasts.

Question 8

The selection of candidate independent variables for a regression model to forecast motel occupancy requires input from knowledgeable marketing personnel.
Multiple Choice

Question 9

Historical analogy involves comparative analysis of the introduction and growth patterns of new items with similar previously-introduced items.

Question 10

Most forecasting models assume that the underlying pattern of behavior of their data will remain the same. The only component of error is attributable to random fluctuations that are not under the control of anyone.

Question 11

Both trend and seasonal adjustments can augment a single exponential-smoothing model.

Question 12

Seasonal adjustment involves the deseasonalization of data in a given cycle after smoothing.

Question 13

The cost of formulating, developing, and testing regression models, plus the expertise required to interpret results, often dictates the use of this model only for medium-term or long-term forecasting.

Question 14

One reason exponential smoothing is so popular is the ease with which it can be made to accommodate trend and seasonality in its forecasts.

Question 15

Subjective models are used to assess the future impact of changing demographics.

Question 16

The main advantage that simple moving-average models have over exponentially-smoothed models is that they can be made to give any desired weight to specific periods in the past.

Question 17

Econometric models are basically regression models that involve a system of equations.

Question 18

In the Delphi method, the opinions of experts are collected, then the analyst resolves differences of opinion to arrive at a consensus.

Question 19

The forecast horizon for a cross-impact analysis method is medium-term.

Question 20

Fast-food restaurants use a causal model to forecast daily demand for menu items.

Question 21

The Delphi method is best used when forecasting:

A) customer demand.
B) technological change.
C) correlations between events.
D) future economic conditions.

Question 22

All but one of the following are characteristics of the moving-average model:

A) very low relative cost.
B) short term forecast horizon.
C) uses all past data in forecast.
D) used for demand forecastsing.

Question 23

Survey results are used in:

A) regression.
B) cross-impact anlaysis.
C) econometric models.
D) Delphi method.

Question 24

____ is the smoothing constant for the trend adjustment in an exponential smoothing forecast.

A) Alpha
B) Beta
C) Gamma
D) Delta

Question 25

The ___________ forecasting method requires the most computer power.

A) econometric
B) cross-impact
C) regression
D) historical analogy

Question 26

Demand for hotel services will require the use of which one of the following forecasting models:

A) subjective and causal.
B) subjective and time series.
C) time series and causal.
D) only time series.

Question 27

Long-term forecasts do not deal with predictions for:

A) new services.
B) changes in mix of services.
C) overtime or temporary labor requirements.
D) process technology.

Question 28

_____________ is used best to forecast demand.

A) Exponential smoothing
B) Historical analogy
C) Dephi method
D) Econometrics

Question 29

Characteristics of forecasting models include all but one of the following:

A) data required.
B) economic conditions.
C) relative cost.
D) forecast horizon.

Question 30

The usual range of the smoothing constant for a simple exponential smoothing forecast is:

A) 0.1-0.5
B) 0.3-0.7
C) 0.5-0.7
D) 0.5-1.0

Question 31

The forecast horizon for regression methods is _________.

A) short term.
B) medium term.
C) long term.
D) medium to long term.

Question 32

In forecasting, the term "cycle" refers to repetition of data:

A) once a year, if seasonal.
B) once in a period.
C) within a week.
D) during a day.

Question 33

Exponential smoothing uses a feedback control mechanism because:

A) $\alpha$ is less than 1.
B) calculations are based on modifying the previous smooth value.
C) the calculation contains the forecast error.
D) the weights given to past data are exponentially distributed.

Question 34

Which one of the following is not an advantage of the Simple Exponential Smoothing model over the N-period Moving Average model?

A) Old data are never dropped.
B) Older data are given progressively less weight.
C) Calculation is simple.
D) None of the above.

Question 35

All of the following are forecasting models, except:

A) subjective.
B) objective.
C) causal.
D) time series.

Question 36

Which oneof the following is an example of a causal model?

A) Delphi method
B) Moving average
C) Cross-impact analysis
D) Regression

Question 37

Regression models require all but one of the following:

A) numerical data.
B) subjective data.
C) independent variables.
D) dependent variable.

Question 38

In regression models, the variable of interest (being forecast) is called a(an):

A) defining variable.
B) independent variable.
C) basic variable.
D) dependent variable.

Question 39

Subjective forecasting models are best characterized by:

A) the use of experts.
B) the use of opinion as a basis for a forecast.
C) the absence of data manipulation.
D) the use of simultaneous equations.

Question 40

The accuracy of a time-series model is dependent upon:

A) the nature of past data.
B) the weight given to past data.
C) the treatment of seasonality and trend.
D) the assumption that the future is a projection of the past.