Forecasting Techniques

At the end of the lesson, you should be

able to:
• define forecasting;
• discuss the importance of forecasting in business;
• identify the types of forecasting;
• explain the approaches of forecasting;
• solve forecasting problems using regression
analysis and trend projection.
 What is Forecasting?
Forecasting is the art and science of
predicting what will happen in the future.
• determined by a mathematical method;
• sometimes it is based on the intuition
of the operations manager.
• most forecasts and end decisions are a
combination of both.
 What is Business Forecasting?
 Business forecasting consists of tools and
techniques used to predict changes in business,
such as sales, expenditures, profits and losses.
The goal of business forecasting is to develop
better strategies based on these informed
predictions; helping to eliminate potential
failure or losses before they happen.
 Why Forecasting Important?
Forecasting is valuable to businesses
because it gives the ability to make
informed business decisions and develop
data-driven strategies.
 Why Forecasting Important?
Financial and operational decisions are made
based on current market conditions and
predictions on how the future looks.
 Why Forecasting Important?
Past data is aggregated and analyzed to find
patterns, used to predict future trends and
changes. Forecasting allows your company to
be proactive instead of reactive.
 Why Forecasting Important?
Good operations managers learn how to
forecast, to trust the numbers, and to trust
their instincts to make the right decisions for
their firm.
Forecasting is conducted by what are referred to
as time horizons.
1. Short range forecast. While it can be up to one
year, this forecast is usually used for three months or
less. It is used for planning purchases, hiring, job
assignments, production levels, and the like.
2. Medium range forecast. This is generally three
months to three years. Medium range forecasts are
used for sales and production planning, budgeting, and
analysis of different operating plans.
3. Long range forecast. Generally three years or more
in time span, it is used for new products, capital
expenditures, facility expansion, relocation, and
research and development.
Medium and long range forecasts differ from
short range forecasts by other characteristics
as well.
1. Medium and long range forecasts are more
comprehensive in nature. They support and
guide management decisions in planning
products, processes, and plants. A new plant
can take seven or eight years from the time it
is thought of, until it is ready to move into
and become functional.
2. Short term forecasts use different
methodologies than the others. Most
short term forecasts are quantitative in
nature and use existing data in
mathematical formulas to anticipate
immediate future needs and impacts.
3. Short term forecasts are more
accurate than medium or long range
forecasts. A lot can change in three
months, a year, three years, and longer.
Factors that could influence those
forecasts change every day. Short term
forecasts need to be updated regularly
to maintain their effectiveness.
 Types of Forecasts
There are three major types of forecasting,
regardless of time horizon, that are used by
organizations.
1. Economic forecasts address the business
cycle. They predict housing starts, inflation rates,
money supplies, and other indicators.
2. Technological forecasts monitor rates of
technological progress. This keeps organizations
abreast of trends and can result in exciting new
products. New products may require new facilities
and equipment, which must be planned for in the
 Types of Forecasts
3. Demand forecasts deal with the
company's products and estimate
consumer demand. These are also
referred to as sales forecasts, which
have multiple purposes. In addition to
driving scheduling, production, and
capacity, they are also inputs to
financial, personnel, and marketing
future plans.
 Forecasting System
These seven steps can generate forecasts.
1. Determine what the forecast is for.
2. Select the items for the forecast.
3. Select the time horizon.
4. Select the forecast model type.
5. Gather data to be input into the model.
6. Make the forecast.
7. Verify and implement the results.
 Forecasting Approaches
There are two predominant approaches to
forecasting:
Qualitative approach uses factors
such as experience, instinct and
emotion while the quantitative
analysis relies heavily on
mathematics, historical data and
casual variables.
 Forecasting Approaches
Qualitative methods include:
1. Jury of executive opinion. This is
based on the inputs and decisions of
high-level experts or management.
2. Delphi method. Decision makers,
staff, and respondents all meet to
develop the forecast. Every shareholder
in the process provides input.
 Forecasting Approaches
3. Sales force composite. Each sales
person provides an individual estimate
which is reviewed for realism by
management, and then combined for a big
picture view.
4. Consumer market survey. This is
surveying the prospective customer base to
determine demand for existing products and
can also be used for new products.
 Forecasting Approaches
Categories Quantitative methods:
 
1. Associative models uses similar historical
data inputs and then includes other external
variables such as advertising budget, housing,
competitor's prices and more (Regression
Analysis).
2. Time-series models predict by assuming the
future is a function of the past (Trend
projection). 
 Simple Linear Regression
Simple linear regression is a statistical method that
allows us to summarize and study relationships between
two continuous (quantitative) variables:
 One variable, denoted x, is regarded as
the predictor, explanatory, or independent variable.
 The other variable, denoted y, is regarded as
the response, outcome, or dependent variable.
 Simple Linear Regression
•Simple linear regression gets its adjective
"simple," because it concerns the study of
only one predictor variable. In contrast,
multiple linear regression, which we study
later in this course, gets its adjective
"multiple," because it concerns the study of
two or more predictor variables.
 Simple Linear Regression
•Simple linear regression gets its adjective
"simple," because it concerns the study of
only one predictor variable. In contrast,
multiple linear regression, which we study
later in this course, gets its adjective
"multiple," because it concerns the study of
two or more predictor variables.
Regression Analysis Model
The job of developing a mathematical
equation can be quite complex, because we
need to have some ideas about the nature of
the relationship between each of the
independent variables and the dependent
variables. The number of different
mathematical models that could be proposed
is virtually infinite.
Regression Analysis Model
• 
Profit = (Price per unit - variable cost per
unit) x number of unit sold – Fixed cost
F = P where F = Future value
P = Present- value
i = interest rate per period
n = number of periods
Regression Analysis Model
These are examples of deterministic
models because such equations allows us
to determine the value of the dependent
variable from the values of the
-

independent variable. For many practical

applications of interest to us deterministic
models are unrealistic.
Deterministic Regression Model

-
Regression Analysis Model
•Note that the observed (x, y) data points
fall directly on a line. As you may
remember, the relationship between
degrees Fahrenheit and degrees Celsius
is known to be:
•F=95C+32 F=95C+32
Simple Linear Regression Model
 Height and weight — as height increases,
you'd expect weight to increase, but not
perfectly.
 Alcohol consumed and blood alcohol content
— as alcohol consumption increases, you'd
expect one's blood alcohol content to increase,
but not perfectly.
Simple Linear Regression Model
 Vital lung capacity and pack-years of smoking — as
amount of smoking increases (as quantified by the
number of pack-years of smoking), you'd expect lung
function (as quantified by vital lung capacity) to
decrease, but not perfectly.
 Driving speed and gas mileage — as driving speed
increases, you'd expect gas mileage to decrease, but
not perfectly.
Simple Linear Regression Model
We must include in most practical models
is a method to represent randomness that
is part of real life process. This model is
called probabilistic model. To create
probabilistic model, let us start with the
deterministic model that approximates the
relationship we want to model.
Simple Linear Regression Model
The real estate agent knows that the cost
of building new house is about P2,000 per
square foot and that most lots sell for
about P200,000. Approximately, the
selling price would be
y = 2,000x + 200,000; if x = 1,000 square
feet, then the y = P2,200,000
Simple Linear Regression Model
In reality the SP ranges from 2M to 3M, in
other words the deterministic model is not
really suitable in many situation properly,
that is why we should use the probabilistic
model
y = 2,000x + 200,000 + ε, where ε is a greek
letter word epsilon that represent error
variable.
Simple Linear Regression Model
Error variable is the difference between
the actual selling price and the estimated
selling price based from size of the house.
So it is very important to consider the
other variables such as the number of
rooms, the location and other variables.
Simple Linear Regression Model
•First
  order linear model or simple linear regression
model
y=
= y-intercept
= slope
= error variable
y = dependent variable
x = independent variable
Simple Linear Regression Model
• 
Least Squares method
=
= y-intercept
= slope
y = dependent variable
x = independent variable
Simple Linear Regression Model
• The annual bonuses in (S1,000) of the six employees with different
years of experiences were recorded as follows. We wish to determine
the straight line relationship between the annual bonus and years of
experience.
Years of experience x Annual bonus xy
y

Σ Σ Σ
• 
Solve for the covariance of x and y

7.5
•= 
=
= 3.5
Solve for
=
=
= 2.114
• 
= y̅ - x̅  
x̅  = = = 3.5
y̅ = = = 8.333
= 8.333 -  
= 0.934
=
=
Simple Linear Regression Model
• The annual bonuses in (S1,000) of the six employees with different
years of experiences were recorded as follows. We wish to determine
the straight line relationship between the annual bonus and years of
experience.
Years of experience x Annual bonus xy
y

Σ Σ Σ
 Scatter Diagram of the Relationship Between Years of Experience and Annual Bonus
18

16

14
f(x) = 2.33 x
12
Annual bonus

10

6 y1 – y1

0
0 1 2 3 4 5 6 7

Years of experience
Advertising in thousand of pesos Sales in million of pesos
23 9.6
46 11.3
60 12.8
54 9.8
28 8.9
33 12.5
25 12.0
31 11.4
36 12.6
88 13.7
90 14.4
99 15.9
• References
• The Art and Science of Forecasting in Operations Management
• https://www.universalclass.com/articles/business/the-art-and-science
-of-forecasting-in-operations-management.htm
• ington Hall | Oct 21, 2020 12:57:16 PM | CRM, Business Intelligence, 
BI, Customer Relationship Management
• https://www.baass.com/blog/why-forecasting-is-important-for-busine
ss-success