Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

COMM225 > Lesson 3: Demand Forecasting > Flashcards

Lesson 3: Demand Forecasting Flashcards

1
Q

What is forecasting?

A

Forecast is an estimate of “something” over a future time period.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the 2 approaches to forecasting?

A

Quantitative and Non-Quantitative (Judgemental)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is the difference between quantitative and non-quantitative approaches to forecasting?

A

Quantitative techniques use data whereas non-quantitative use analysis of subjective inputs.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

How is the judgemental method implemented?

A

It is based on opinions. For experts, they may look at past experiences whereas a sales force team might look to their customers’ opinions.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What are ways to gather information for the judgemental method?

A

Consumer surveys and historical analogies (using demand for a similar product)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is the Delphi method?

A

The Delphi method is a widely popular approach to develop consensus on the experts opinions. Opinions are listed in the questionnaire format and, with the help of a coordinator and discussions, questionnaire is updated iteratively (by eliminating non supportive opinions) till full consensus is reached.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is time series?

A

a time ordered sequence of observations taken at regular intervals of time.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What are the six patterns in a time series?

A
  • Level
  • Trend
  • Seasonality
  • Cycles
  • Irregular Variations
  • Random Variation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is “level” (in time series)?

A

(Average) Horizontal Pattern

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is “trend” (in time series)?

A

Steady upward or downward movement

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is “seasonality” (in time series)?

A

Regular variations related to time of year r day

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is “cycles” (in time series)?

A

Wavelike variations lasting more than one year

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is “irregular variations” (in time series)?

A

Caused by unusual circumstancces, not reflective of typical behavior

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is “random variations” (in time series)?

A

Residual variations after all other behaviors are accounted for (called noise)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What are the methods to calculate time series corrections?

A
  • Naive Method
  • Moving Average
  • Weighted Moving Average
  • Exponential Smoothing
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Naive Method:

A

Next period = last period demand

It is simple to use and understand. It has low cost and low accuracy.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Moving Average:

A

In the Moving Average, we take the average of a few recent demands. If we take the average of three recent demands, we call it the 3 period moving average.

If we take 5, it’s 5 day moving average.

To summarize, the Moving Average is the average of the last few actual data values, which is updated each period.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Weighted Moving Average:

A

In the Moving Average Method, we give equal weight to each period. But if we have some idea about the trend, business environment, or political situation/announcement, we can give a different weight to each data.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Exponential Smoothing:

A

Exponential smoothing is also a weighted moving average method. However, in this method, we give some weight to previous forecast too.
New forecast is based on the previous forecast plus a percentage of the difference between that forecast and the previous actual value.
New Forecast equals the previous forecast plus a percentage of the forecast error.
(Actual – Forecast) is the error term. Alpha is the percentage of the error applied to the previous forecast to generate a new forecast.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What are the types of trends?

A

Nonlinear and linear

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What is a nonlinear trend?

A

Non linear trends can take various shapes and thus difficult to estimate i.e. formulate in a form of formula.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What is a linear trend?

A

Linear trends are easy to estimate because they are like a line. Non linear trends can take various shapes and thus difficult to estimate i.e. formulate in a form of formula.

A linear trend line develops a “line of best fit” (tread line) amongst the actual observation “data points”.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

What is trend adjusted smoothing?

A

A variation of simple exponential smoothing can be used when a time series exhibits a trend. It is called

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What are seasonal variations?

A

Seasonal Variations: regularly repeating wavelike movements in series values that can be tied to recurring events, weather, or a calendar. Examples of seasonality are retail trade, ice cream production, and residential natural gas sales.

Most seasonal variations repeat annually, and are applied to shorter lengths of repeating patterns: Rush hour traffic occurs twice a day. Theatres and restaurants demands are higher on Fridays or on weekends. Banks may experience daily and weekly repeating “seasonal” variations (heavier traffic at lunch, just before closing, on Friday)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

What are the two models of trend-adjusted exponential smoothing?

A

Additive: Quantity added to average or trend.

Multiplicative: Proportion * average or trend

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

What is seasonal relative (index)?

A

Equals proportion of average or trend for a season in the multiplicative model. Seasonal relative of 1.2 = 20% above average.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

What is deseasonalize?

A

Remove seasonal component to more clearly see other components. Divide by seasonal relative.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

What is reasonalize?

A

Adjust the forecast for seasonal component. Multiply by seasonal relative.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

Time Series Decomposition:

A
  1. Compute the seasonal relatives.
  2. De-seasonalize the demand data.
  3. Fit a model to de-seasonalized demand data, e.g., moving average or trend.
  4. Forecast using this model and the de-seasonalized demand data.
  5. Re-seasonalize the deseasonalized forecasts.

In a simple term, first we remove the wave like pattern from the data so that data looks stable or with trends and then use our basic models to forecast. And finally, we correct our forecast.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
30
Q

What do associative models rely on?

A

Associative models rely on identification of related variables that can be used to predict values of the variable of interest.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
31
Q

What are predictor variables (x)?

A

Used to predict values of the variable of interest (y). - Also called independent variables.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
32
Q

What is linear regression?

A

Process of finding a straight line that best fits a set of points on a graph. - Use the Least Square Equation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
33
Q

What is multiple regression?

A

Models with more than one predictor variable. Computations complex -> created with computer.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
34
Q

What do you use for regression analysis?

A

Linear regression (generally the most popular); Least square technique formula; Causal variables (multiple regression)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
35
Q

What are the forecasting errors in time series analysis?

A
  1. Projection of past trends into the future.
  2. Bias errors.
  3. Random errors.
36
Q

What are bias errors?

A

Errors due to some mistakes, wrong parameters, etc.

37
Q

What are random errors?

A

Errors that cannot be explained or something that cannot be modeled i.e. randomness or random error.

38
Q

What is mean absolute deviation (MAD)?

A

Average forecasting error based on the absolute difference between actual and forecast demands.

39
Q

What is mean absolute percentage error (MAPE)?

A

Used to determine the forecasting errors as a percentage of the actual demand.

40
Q

What is mean squared error (MSE)?

A

A measure of the variability in the forecasting error.

41
Q

What is tracking signal?

A

Used to track the accuracy of the forecasts. Tracking signal suggests the need of recalibrating the forecasting model.

Positive tracking signal means “Actual consistently exceeds forecast”.

Negative tracking signal means “Actual is consistently below the forecast”.

42
Q

True or False:
Forecasting techniques generally assume that the same causal system that existed in the past will continue to exist in the future.

A

True

43
Q

True or False:

Forecasts are rarely perfect.

A

True

44
Q

True or False:
Generally, the responsibility for preparing demand forecasts for finished goods or services lies with operations rather than marketing or sales departments.

A

False

45
Q

True or False:
Forecasts for groups of items tend to be less accurate than forecasts for individual items because forecasts for individual items are not subject to as many influencing factors.

A

False

46
Q

True or False:
Organizations that are capable of responding quickly to changing requirements can use a shorter forecast horizon and therefore benefit from more accurate forecasts.

A

True

47
Q

True or False:

Forecast accuracy tends to increase as the time horizon increases.

A

False

48
Q

True or False:
The purpose of the forecast should be established first so that the level of detail, amount of resources, and accuracy level can be indicated.

A

True

49
Q

All of the following are true about forecasts, except:

  • They are always perfect
  • They are less accurate for individual items than for group items
  • They are usually the responsibility of operating managers to prepare
  • They become less accurate with longer time horizons
  • They assume the same underlying causal system in the future as the past
A

They are always perfect

50
Q

A company is conducting long-term planning of which types of services they should offer. Which of the following forecasting techniques are they most likely to use?

  • Simple exponential smoothing
  • Executive opinion
  • Naive method
  • Trend models
  • Regression models
A

Executive opinion

51
Q

A managerial approach toward forecasting which seeks to actively influence demand is:

  • Reactive
  • Retroactive
  • Proactive
  • Reflexive
  • Protracted
A

Proactive

52
Q

Which of the following techniques are most likely to be used for forecasting demand for new products and services?

  • Regression models
  • Exponential smoothing techniques
  • Moving averages
  • Judgemental methods
  • Trend models
A

Judgemental methods

53
Q

The two most important factors in choosing a forecasting technique are:

  • Cost and ease of use
  • Accuracy and buy-in
  • Accuracy and time horizon
  • Cost and time horizon
  • Cost and accuracy
A

Cost and accuracy

54
Q

True or False:

Forecasts based on time series (historical) data are referred to as associative forecasts.

A

False

55
Q

True or False:
Forecasting techniques that are based on time series data assume that future values of the series will duplicate past values.

A

False

56
Q

True or False:
Increasing the number of data points included in a moving average will result in a forecast that is smoother but less responsive to changes.

A

True

57
Q

True or False:

The naive forecast is limited in its application to series that reflect no trend or seasonality.

A

False

58
Q

True or False:
A moving average forecast tends to be more responsive to changes in the data series when more data points are included in the average.

A

False

59
Q

True or False:

A simple moving average assigns equal weight to each data point that is represented by the average.

A

True

60
Q

True or False:
An advantage of a weighted moving average is that more recent experience is given more weight than less recent experience.

A

True

61
Q

True or False:

Exponential smoothing is a form of weighted averaging.

A

True

62
Q

True or False:
A smoothing constant of 0.1 will cause an exponential smoothing forecast to react more quickly to a sudden change than a value of 0.3 will.

A

False

63
Q

True or False:
In exponential smoothing, an alpha of 0.30 will cause a forecast to react more quickly to a large error than will an alpha of 0.20.

A

True

64
Q

True or False:

The forecast error is the difference between the actual value and the forecast value for a given period.

A

True

65
Q

True or False:

MAD is equal to the square root of MSE.

A

False

66
Q

True or False:
The MSE is the best measure to use in a control chart to monitor if forecast error is randomly distributed around a mean value of 0.

A

False

67
Q

Which of the following is not an accurate statement concerning bias in forecasts?

  • Bias is calculated based on the mean absolute percent error (MAPE)
  • Bias is the sum of forecast errors
  • Persistent negative bias implies forecasting frequently overstating actual values
  • Bias may indicate a shift in the demand pattern
  • Persistent positive bias implies frequently underestimating actual values
A

Bias is calculated based on the mean absolute percent error (MAPE)

68
Q

Given forecast errors of 5, 0, -4, and 3, what is the mean absolute deviation (MAD)?

  • 3
  • 1
  • 2.5
  • 2
  • 4
A

3

69
Q

Positive forecast errors mean that the forecast:

  • Was too low
  • Was irregular
  • Is where the predictor variable indicated
  • Was too high
  • Was accurate
A

Was too low

70
Q

MSE weighs errors according to ______________ and MAPE weighs according to _______________.

  • Absolute error; average error
  • Absolute values; absolute percentage error
  • Squared values; mean absolute values
  • Squared values; absolute percentage error
  • Absolute percentage error; squared values
A

Squared values; absolute percentage error

71
Q

Which is not a characteristic of exponential smoothing?

  • Weighs each historical value equally
  • Smooths random variations in the data
  • Provides an easily altered weighting scheme
  • Directly accounts for forecast error
  • Smooths real variations in the data
A

Weighs each historical value equally

72
Q

Which of the following smoothing constants would make an exponential smoothing forecast equivalent to a naive forecast?

  • 0.01
  • 5
  • 0
  • 1
  • 1.0
A

1.0

73
Q

As compared to a simple moving average, the weighted moving average is:

  • Easier to compute
  • Less reflective of the most recent periods
  • Smoother
  • More reflective of the most recent periods
  • More readily able to identify random variations
A

More reflective of the most recent periods

74
Q

In order to increase the responsiveness of a forecast made using the moving average technique, the number of data points in the average should be:

  • Multiplied by a larger alpha
  • Divided by alpha
  • Decreased
  • Multiplied by a smaller alpha
  • Increased
A

Decreased

75
Q

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

  • 0
  • 0.15
  • 0.01
  • 0.05
  • 0.10
A

0.15

76
Q

True or False:

The naive approach to forecasting requires a linear trend line.

A

False

77
Q

True or False:

Time series techniques involve identification of explanatory variables that can be used to predict future demand.

A

False

78
Q

True or False:

A seasonal relative (or seasonal indexes) is expressed as a percentage of the average or trend in a time series.

A

True

79
Q

True or False:

In order to compute seasonal relatives, the trend of past data must be computed or known.

A

True

80
Q

True or False:
Removing the seasonal component from a data series (deseasonalizing) can be accomplished by dividing each data point by its appropriate seasonal relative.

A

True

81
Q

Persistent upward or downward movement in time series data is called:

  • Seasonal variation
  • Irregular variation
  • Random variation
  • Trend
  • Cycles
A

Trend

82
Q

The primary difference between seasonality and cycles is:

  • The ability to attribute the pattern to a cause
  • There is less forecasting “noise” in a cycle
  • There is more forecasting “noise” in a cycle
  • The magnitude of the variation
  • The duration of the repeating patterns
A

The duration of the repeating patterns

83
Q

Which of the following are most likely to be used for forecasting demand for the longer term?

  • Simple exponential smoothing
  • Judgemental methods
  • Delphi method
  • Regression trend models
  • Naive method
A

Regression trend models

84
Q

True or False:

Forecasts based on time series (historical) data are referred to as associative forecasts.

A

False

85
Q

True or False:
Multiple regression procedures permit us to extend associative models to relationships that involve more than one predictor variable.

A

True

86
Q

The primary method for associative foecasting is:

  • Exponential smoothing
  • Simple moving averages
  • Centred moving averages
  • Naive method
  • Regression analysis
A

Regression analysis

87
Q

Which term most closely describes what associative forecasting techniques are based on?

  • Predictor variables
  • The Delphi technique
  • Linear relationships
  • Consumer survey
  • Time series data
A

Predictor variables

COMM225 (8 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions