You are on page 1of 49

Managerial Economics

Applications, Strategies and Tactics, 14e

James R. McGuigan
R. Charles Moyer
Frederick H. deB. Harris

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a
1
license distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
PART II – DEMAND AND FORECASTING

Chapter 5 –
Business and Economic Forecasting

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
2
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Chapter 5 – Business and Economic Forecasting
Overview (1 OF 2)
• THE SIGNIFICANCE OF FORECASTING
• SELECTING A FORECASTING TECHNIQUE
• ALTERNATIVE FORECASTING TECHNIQUES
• DETERMINISTIC TREND ANALYSIS
• SMOOTHING TECHNIQUES
• BAROMETRIC TECHNIQUES
• SURVEY AND OPINION-POLLING TECHNIQUES
© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
3
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Chapter 5 – Business and Economic Forecasting
Overview (2 OF 2)

• ECONOMETRIC MODELS
• FORECASTING WITH INPUT-OUTPUT TABLES
• STOCHASTIC TIME-SERIES ANALYSIS

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
4
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Figure 5.1 – Market Penetration Curves of the
Color Television and the Internet (See Managerial Challenge)

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
5
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – The Significance of Forecasting (1 of 1)

• Accurately forecasting future business prospects is one of the


most important functions of management
• Sales forecasts are necessary for operations managers to plan
future levels of production
• Financial managers require estimates of future sales revenue,
disbursements and capital expenditures
• Forecasts of credit conditions will direct the cash needs of the
firm at the lowest cost
• Public administrators and managers must similarly forecast
© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
6
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Selecting a Forecasting Technique (1 of 2)

• Hierarchy of Forecasts
• The highest level is national, the gross domestic product (GDP), tho a
firm may be interested in some specific component of GDP
• Next are industry sales forecasts; then individual firm sales forecasts
• Within the firm, another hierarchy of forecasts exists:
• Managers estimate company-wide or regional dollar sales and unit
sales by product line for the use of operations, marketing, and sales
• Long term forecasts for the economy, industry and firm are used in
planning long-term capital expenditures, and for charting the firm’s
direction

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
7
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Selecting a Forecasting Technique (2 of 2)

• Criteria Used to Select a Forecasting Technique


• 1. the cost associated with developing the forecasting model
• 2. the complexity of the relationships that are being forecast
• 3. the time period of the forecast (long-term or short-term)
• 4. the lead time needed to make decisions based on the forecast, and
• 5. the accuracy required of the forecasting model
• Evaluating the Accuracy of Forecasting Models

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
8
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
What Went Right? ● What Went Wrong?
Crocs Shoes
• An overnight sensation, in 2002, a colorful foam clog, lightweight and
nearly indestructible appeared on the market; 100 million pairs were sold
in 7 years.
• Forecasting double-digit sales growth, the company’s IPO was successful;
the capital was invested in manufacturing capacity.
• The worldwide recession of 2008-09 caused the bottom to fall out of the
market; no one needed replacements for a nearly indestructible fashion
fad that couldn’t help look for a job.
• A profit of $168 million became a loss of $185 million.
• After several years of retrenchment, Crocs are now profitable, but at less
than one-tenth of their previous capacity.
© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
9
Ch 5 – Alternative Forecasting Techniques (1 of 1)

• There are a range of forecasting techniques:


• 1. Deterministic trend analysis
• 2. Smoothing techniques
• 3. Barometric indicators
• 4. Survey and opinion-polling techniques
• 5. Macroeconometric models
• 6. Stochastic time-series analysis
• 7. Forecasting with input-output tables

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
10
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Deterministic Trend Analysis (1 of 1)
• Data collected for use in forecasting the value of a particular
variable may be classified as:
• Time-series – a series of observations taken on an economic
variable at various points in time
• Cross-sectional data – Series of observations taken on different
observation units (e.g., households, states, or countries) at the
same point in time

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
11
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Deterministic Trend Analysis
Components of a Time Series (1 of 1)

• Secular trends – long-run changes (growth or decline) in an


economic time-series variable
• Cyclical variations – major expansions and contractions in an
economic series that usually are longer than a year in duration
• Seasonal effects – variations in a time series during a year that
tend to appear regularly from year to year
• Random fluctuations – factors that are unpredictable, such as
hurricanes, floods, extraordinary government actions like a wage-
price freeze or declaration of war (See Figure 5.2)
© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
12
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Figure 5.2 – Secular, Cyclical, Seasonal, and
Random Fluctuations in Time-Series Data

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
13
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Deterministic Trend Analysis
Some Elementary Time-Series Models (1 of 2)
• The simplest time-series model states that the forecast value of the
variable for the next period will be the same as for the present
period; see Table 5.1

• This model may be useful where change occurs slowly and the
forecast is made for a relatively short period in the future, but
• The forecaster may need to speed up the collection of actual data
• The model makes no provision for incorporating special promotions by
the firm
© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
14
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Deterministic Trend Analysis
Some Elementary Time-Series Models (2 of 2)
• The tendency for recent increases to trigger further increases in
beer sales may be incorporated by adjusting equation 5.2:

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
15
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Table 5.1 – Buckeye Brewing Company’s
Monthly Beer Sales (Thousands of Barrels)

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
16
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Deterministic Trend Analysis
Secular Trends (1 of 4)
• Long-run changes in an economic time series can follow different
types of trends; three possible are shown in Figure 5.3:
• Panel (a) shows a linear trend
• Panel (b) shows a nonlinear constant rate of growth pattern
• Panel (c) depicts a nonlinear trend of a declining rate of growth

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
17
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Figure 5.3 – Time-Series Growth Patterns

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
18
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Deterministic Trend Analysis
Secular Trends (2 of 4)
• Linear Trends – A linear time trend may be estimated by using
least-squares regression analysis to provide an equation of a
straight line of “best fit”

• Linear time trend forecasting is easy and in expensive, but too


simple and inflexible to be used in many forecasting circumstances

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
19
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Figure 5.4 – Prizer Creamery:
Monthly Ice Cream Sales

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
20
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Deterministic Trend Analysis
Secular Trends (3 of 4)
• Constant Rate of Growth Trends– The formula for the constant rate
of growth forecasting model is:

• Because Equation 5.5 is a nonlinear relationship, the parameters


cannot be estimated directly with the ordinary least-squares
method. But taking logarithms of both sides of the equation gives:

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
21
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Deterministic Trend Analysis
Secular Trends (4 of 4)
• Declining Rate of Growth Trends– The curve shown in Figure 5.3
Panel © is particularly useful for representing sales penetration
curves in marketing applications
• Using linear regression techniques, one can specify a semilog
estimating equation:

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
22
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Deterministic Trend Analysis
Seasonal Variations (1 of 1)
• Seasonal variations may be estimated in a number of ways:
• Ratio-to-Trend Method – This method assumes that the trend value is
multiplied by the seasonal effect
• Dummy Variables – This is a variable that normally takes on one of two
values – either 0 or 1 – and is used to capture the impact of certain
qualitative factors in an econometric relationship; assumes that seasonal
effects are added to the trend value

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
23
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Table 5.2 – Prizer Creamery’s August Ice
Cream Sales

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
24
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Figure 5.5 – Actual and Expected Inflation

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
25
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Smoothing Techniques (1 of 1)

• Smoothing techniques are a type of forecasting model which


assumes that a repetitive underlying pattern can be found in the
historical values of a variable being forecast
• Moving Averages
• A series of recent observations can be averaged to arrive at a forecast; observed
values are chosen, their average is computed, and serves as a forecast:

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
26
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Table 5.3 – Value-Mart’s Quarterly Sales
Forecast (2008)

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
27
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Table 5.4 – Walker Corporation’s Three-Month
Moving Average Sales Forecast Table

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
28
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Smoothing Techniques
First-Order Exponential Smoothing (1 of 4)

• Exponential smoothing is designed to overcome the objection that


smoothing techniques normally give equal weight to all
observations, even though the most recent observation probably
contains more useful information; consider this alternative:

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
29
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Figure 5.6 – Walker Corporation’s Three-
Month Moving Average Sales Forecast Chart

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
30
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Smoothing Techniques
First-Order Exponential Smoothing (2 of 4)

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
31
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Smoothing Techniques
First-Order Exponential Smoothing (3 of4)

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
32
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Smoothing Techniques
First-Order Exponential Smoothing (4 of 4)

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
33
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Table 5.5 – Walker Corporation’s First-Order
Exponential Smoothing Sales Forecast

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
34
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Barometric Techniques (1 of 1)
• Time-series forecasting models assume that future patterns in an
economic time series may be predicted by b y projecting a repeat
of past patterns, but very few economic time series exhibit
consistent enough cyclical variations to make this method reliable;
see Table 5.6
• Leading, Lagging, and Coincident Indicators
• Indicators may be so classified depending on their timing relative
to business cycle peaks and troughs; see Figure 5.7
• Leading and lagging indicators predict the direction of future
change in economic activity; not the magnitude of the changes
© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
35
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Table 5.6 – Duration of U.S. Business Cycles
(In Months)

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
36
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Figure 5.5 – Barometric Indicators

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
37
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Survey and Opinion-Polling Techniques (1 of 2)
• The greatest value of survey and opinion-polling techniques is that
they may help to uncover if consumer tastes are changing, or if
executives begin to lose confidence in the economy
• Forecasting Macroeconomic Activity
• Some of the best known surveys include the following:
• 1. Plant and equipment expenditure plans
• 2. Plans for inventory changes and sales expectations
• 3. Consumer expenditure plans

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
38
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Survey and Opinion-Polling Techniques (2 of 2)
• Sales Forecasting
• These techniques can also be used on a micro level within the firm for
forecasting sales; some of the variations used include the following:
• Sales force polling – some firms poll their own salespeople in the field
about their expectations for future sales by geographical area and
product line; employees closest to customers may have significant
insights into the state of the market
• Surveys of consumer intentions – some firms (especially in durable goods)
conduct their own surveys of specific consumer purchases
• Example: an auto dealer who pursues a “customer for life” relationship with its
target market
© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
39
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Econometric Models (1 of 1)
• Econometrics is a combination of theory, statistical analysis, and
mathematical model building; models can range from simple to
extremely complex
• Advantages of Econometric Forecasting Techniques
• These possess significant advantages over other types of models
• The most significant advantage is that they identify independent variables
that the manager may be able to manipulate
• Another advantage is that they predict not only the direction of change in
an economic series, but also the magnitude of that change

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
40
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Econometric Models
Single-Equation Models (1 of 1)

• The simplest form


• A single-equation model was used in Ch 4 to explain the demand
for Sherwin-Williams house paint
• Once the parameters of the demand equation are estimated, the
model can be used to make forecasts of demand for a particular
product in a given region

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
41
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Econometric Models
Multi-Equation Models (1 of 2)

• When the interrelationships are complex:

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
42
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Econometric Models
Multi-Equation Models (2 of 2)

• Complex Models of the U.S. Economy


• A number of multiple-equation econometric models of the U.S. economy
have been developed and are used to forecast business activity
• Information on three of these models and the forecasting techniques they
employ is summarized in Table 5.7
• Some of these models still rely heavily on the judgment of their staffs of
economists

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
43
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Table 5.7 – Characteristics of Three
Econometric Models of the U.S. Economy

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
44
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Econometric Models
Consensus Forecasts: Livingston & Blue Chip Forecaster Surveys (1 of 1)
• The Federal Reserve Bank of Philadelphia (Livingston surveys) and Blue Chip
Economic Indicators in Aspen, Colorado, conduct semiannual surveys of
leading U.S. economists regarding their forecasts of unemployment, inflation,
stock prices and economic growth
• The 50-60 economists regularly surveyed represent a cross section from
large corporations, banks, labor unions, government, banking firms and
universities
• These forecast surveys have been used by federal and state budget offices
and many corporations to gauge future economic growth & inflation
• These forecasts are more stable over time than any individual forecast
© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
45
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Figure 5.8 – Livingston and Blue Chip
Forecasts of GDP Growth

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
46
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Forecasting with Input-Output Tables (1 of 1)
• Input-Output analysis is another approach that capitalizes on the
cross-sectional interdependence between various intermediate
product and final product industries
• This analysis enables the forecaster to trace the effects of an
increase in demand for one product to other industries
• An increase in the demand for autos will first lead to an increase in the
demand for steel, glass, plastics, tires and upholstery fabric
• Also, secondary impacts will occur as the increase the increase in the
demand for upholstery fabric (i.e., fibers used to make the fabric)

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
47
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Stochastic Time-Series Analysis
Advanced Material (1 of 2)
• Consider these, the two forecasting approaches that capitalize on
the interdependencies in business data
• Stochastic time-series analysis attempts to remove deterministic
time trends and instead model, estimate and hopefully replicate
the stochastic process generating any remaining patterns in the
data across successive time periods (any remaining autocorrelation
patterns)
• Consider:

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
48
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.
Ch 5 – Stochastic Time-Series Analysis
Advanced Material (2 of 2)
• Random walk variables pose several problems if one tries to make
a forecast based on ordinary least-squares (OLS) regression analysis
• Two random walk variables with positive (negative) drift will almost
certainly exhibit spurious correlation
• Also, the level of yT after a number of periods T is:

© 2017 Cengage Learning® May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part, except for use as permitted in a license
49
distributed with a certain product or service or otherwise on a password-protected website or school-approved learning management system for classroom use.

You might also like