Chapter 18: Forecasting

In the exponential smoothing method, how many pieces of information are needed to forecast the future?

3

Linear regression forecasting

A forecasting technique that fits a straight line to past demand data.

At

Actual demand for the period t

Time series analysis

Based on the idea that data relating to past demand can be used to predict future demand.

Forecast errors can be classified as bias or random. Which one can the forecaster reduce?

Bias

errors occur when a consistent mistake is made.

Bias

is the most used of all forecasting techniques.

Exponential smoothing

Strategic forecasts

Medium and long-term forecasts that are used for decisions related to strategy and aggregate demand.

forecasts are useful for capturing seasonal effects.

Medium term

What type of forecasts rely on the knowledge of experts and require judgment?

Qualitative

The Delphi method the identity of the individuals participating in the study.

conceals

In general, the longer the averaging period

the smoother the plot.

One common source that many forecasters are unaware of is projecting past into the future.

trends

Causal forecasting

Assumes that demand is related to some underlying factor or factors in the environment.

of a time series means identifying and separating the time series data into various components.

Decomposition

research is used mostly for product research in the sense of looking for new product ideas.

Market

With market research, the data collection methods are primarily and .

Surveys Interviews

Simulation

These models allow the forecaster to run through a range of assumptions about the condition of the forecast.

In forecasting, there is more than one use for linear regression. What are they?

Time series forecasting Causal relationship forecasting

n

Total number of periods

Which measurement of error can be used to detect forecast bias?

Tracking Signal

All forecasts certainly contain some

error

Typically, fairly small values are used for alpha and delta in the range of what?

0.1 to 0.3

Linear regression

10 to 20 observations

Which of the following equations calculates the MAPE?

100n100n[Σ∣∣At−Ft∣∣At]

If a firm produced a product where they were expecting growth, what alpha value would you expect them to use?

15 to 30

Trend and seasonal models

2 to 3 observations per season

If a firm produced a standard item with relatively stable demand, what alpha value would you expect them to use?

5 or 10

Weighted moving average and simple exponential smoothing + exponential smoothing with trend

5 to 10 observations needed to start

Simple moving average

6 to 12 months; weekly data are often used

Tracking signal

A measure of whether the forecast is keeping pace with any genuine upward or downward changes in demand. This is used to detect forecast bias.

∣∣

A symbol used to indicate the absolute value disregarding positive and negative signs

Exponential smoothing

A time series forecasting technique using weights that decrease exponentially (1 − α) for each past period.

With exponential smoothing, which smoothing constant controls the speed of reaction to differences between forecasts and actual demand?

Alpha

What is the first step in CPFR?

Creation of a front-end partnership agreement

Which of the following is not a forecast horizon used in forecasting?

Current

What term describes when you identify and separate time series data into components?

Decomposition

Which qualitative technique develops a forecast by polling participants multiple times on expected forecast?

Delphi method

Point-of-sale (POS) data would be used in which step of the CPFR process?

Development of demand forecasts

When decisions in forecasting are at a broader, higher level (as when introducing a new product line or concerning strategic product decisions such as new marketing areas), the term is generally used.

Executive judgement

With weighted moving averages, how are the weights selected? Check all that apply.

Experience Trial and error

When the most recent occurrences are more indicative of the future than those in the distant past, one should use which time series model?

Exponential

Ft

Forecast demand for the period t

Causal relationship forecasting

Forecasting using independent variables other than time to predict future demand.

Which of the following is not a reason why exponential smoothing has been well accepted?

Formulating the model is relatively difficult

Which of the following is the formula for a simple moving average?

Ft=At−1+At−2+At−3+...+At−n/n

Which of the following is the formula for a exponential smoothing with trend?

Ft=FITt-1+α(At−1−FITt−1)

Which of the following is the formula for a single exponential smoothing?

Ft=Ft-1+α(At−1−Ft−1)

Which of the following is not an example of a qualitative forecasting technique?

Gamma method

When choosing weights for a weighted moving average, the general rule states that the most recent periods should get the ________________ weighting.

Highest

When releasing a new product that is similar to an existing product, one could use which qualitative technique to develop a forecast?

Historical analogy

What is the major restriction in linear regression forecasting?

It assumes that projections will fall in a straight line

Which measure of error calculates the average absolute value of the actual forecast error?

MAD

Which measurement of error measures the dispersion of some observed value from some expected value?

Mean absolute deviation

In business forecasting, what time frame usually refers to three months to two years?

Medium

When considering more than one variable, along with the effects of each item on the item of interest, which forecasting model should be used?

Multiple regression

Which qualitative forecasting technique is based upon the idea that a group of people with various backgrounds can develop a more reliable forecast than a narrower group?

Panel Consensus

Which of the following is not a component of demand?

Planned

A moving average can be useful in removing what for forecasting.

Random fluctuations

In statistics, the difference between what actually occurred and what was forecast is referred to as what?

Residuals

What term refers to the amount of correction needed in a time series to adjust for the time of the year?

Seasonal factor

models are especially good for measuring the current variability in demand.

Short term

In general, the - models compensate for random variation and adjust for changes over a brief period of time.

Short-term

As the moving average period becomes and periods are used, there is more oscillation and there is a closer following of the trend.

Shorter and fewer

What type of forecast should be used for strategy, sourcing and location decisions?

Strategic

You generally use short-term forecasts for decisions such as replenishing inventory or scheduling employees in the near term.

Tactical

Qualitative

Techniques that use managerial judgment.

At-1

The actual demand in the prior period

Mean absolute percent error (MAPE)

The average error measured as a percentage of average demand.

Mean absolute deviation (MAD)

The average of the absolute value of the actual forecast error.

α

The desired response rate, or smoothing constant

Forecast error

The difference between actual demand and what was forecast.

Ft

The exponentially smoothed forecast for period t

Ft-1

The exponentially smoothed forecast made for the prior period

Smoothing constant alpha (α)

The parameter in the exponential smoothing equation that controls the speed of reaction to differences between forecasts and actual demand.

Decomposition

The process of identifying and separating time series data into fundamental components such as trend and seasonality.

The smoothing constant values depend on which of the following?

The variation in the demand How steady the trend factor is

When you are trying to determine whether a forecast is keeping pace with changes in demand, you should use which measure?

Tracking signal

Which of the following is not a reason why exponential smoothing has been well accepted?

Which of the following is not a reason why exponential smoothing has been well accepted?

Exponential smoothing with trend requires three equations. Which of the following is not one of those three equations?

Yt=a+bt

Exponential smoothing forecasting models are surprisingly

accurate

The two types of seasonal variation discussed are and

additive multiplicative

The main disadvantage of the moving average is that

all individual data elements must be carried as data.

With a simple moving average, the idea is to simply calculate the demand over the most recent periods.

average

Which type of seasonal variation assumes that the season amount is a constant?

Additive

t

Period number

variations are caused by chance events.

Random

Long-term forecasting models detect general trends and are especially useful in identifying major

turning points

In trying to forecast demand for a new product, an ideal situation would be one where an existing product or generic product could be used as a

model

In time series analysis, ____________ demand is used to predict __________ demand.

past; future

The values selected for alpha and delta depend on how much variation there is in demand and how the trend factor is.

random steady

Tactical forecasts are _________ term, while strategic forecasts are _____________ term.

short; medium and long

The exponential smoothing forecasting model uses the following data, except:

smoothing constant delta

Which of the following equations calculates the MAD?

Σ∣∣At−Ft∣∣n

In calculating the MAPE, which of the following terms is not used?

α

Exponential smoothing with trend uses two constants. What are they?

δ α