CHAPTER

TWO
FORECASTING
 INTRODUCTION
Every day manager make a decision without knowing what will be happen in the future.
They order the inventory which out knowing about
 what the sale will be
 what about future further demand
 what about the future technology and other factor

 Managers are always trying to make better estimate of what will happen in the
future in the face of uncertainty. Making good estimates is the main purpose of
forecasting
 forecasting the art and science of predicting future event.
 It involves estimation of the occurrence, timing, and/or magnitude of uncertain
future events or levels of activities.
Forecasting may involve taking historical data and projecting them into the future
with some sort of mathematical model.
 CONT’D
 Successful forecasting requires
Experience
 judgment, and
technical expertise will all play a role in successful forecasting.
 COMMON FEATURES TO ALL FORECASTS
There are wide verity method of forecasting in use, and quite different
from each other, nevertheless there are some common features
 Forecasting techniques generally assume that same underlying causal
system that existed in the past will continue to exist in the future.
 Forecasts are rarely perfect: actual results will usually differ from predicted
values.
 Forecasts for groups of items tend to be more accurate than
forecasts for individual items because forecasting errors among
items in a group usually have a canceling effect.
Forecast accuracy decreases as the time period covered by the
forecast (i.e. the time horizon) increases.
 ELEMENTS OF A GOOD FORECAST

The forecast should be timely

The forecast should be accurate and the degree of accuracy should be stated
The forecast should be reliable (consistent)
The forecast should be expressed in meaningful terms. Different users may
need different terms (Financial in dollars, marketing in units by products,
operations by units, human resources by skills, etc)
The forecast should be in writing
The forecasting technique should be simple to understand and use.
 FORECASTING TIME HORIZONS

•A forecast is usually classified by the future time horizon that it covers. Time horizon
falls in to three categories:

1. Short- range forecast:- this forecast has a time span which is less than three
months. It is used for planning purchasing, job scheduling, workforce levels, job
assignments, and production levels.

2. Medium- range Forecast :- it is intermediate forecast it spans from 3 months to 3 years. It

is useful in sales planning, production planning and budgeting, cash budgeting, and
analyzing various operating plans.
3. Long- range forecast:- Generally 3 years or more in time span, long-range forecast are used in
planning for new products, capital expenditures, facility location or expansion and research
and development.
 STEPS IN THE FORECASTING PROCESS

Step one # determine the purpose of the forecast and when it will be needed.
this will provide an indication of the level of detail required in the forecast, the
amount of resource that can be needed, and the level of accuracy necessary,
# Step Two Establish the time horizon that the forecast must cover.
# Step Three Select the forecast technique
# Step Four Gather and analyze the appropriate data and then prepare the
forecast.
# Step five make the forecast
# Step six monitor the forecast to see if it is performance in a satisfactory
manner. If it is not, re examine the method, assumptions and validity of the
data, and modify as needed, and prepare a revised forecast .
 METHODS /APPROACHES TO
FORECASTING
 Generally there are two basic approaches for forecasting, qualitative and
quantitative.
 Quantitative forecasts use a variety of mathematical models that relay
on historical data and/or causal variables to forecast demand.
 Qualitative forecasts incorporate such variables as the decision maker’s
intuition, emotions, personal experiences, and value system in reaching a
forecast. These factors are often omitted or down played when
quantitative techniques are used because they are difficult or impossible
to quantify.
QUALITATIVE FORECASTING APPROACHES
 It is subjective or judgmental in nature and based on
estimate and opinions.
It is more important/ applicable if
– if forecast must be prepared quickly and not have enough time to
collect and analyze the data.
– If the political and economic conditions are changing
– If the data are obsolete and updated information is not available
– If forecast for new product or new market
TYPE OF QUALITATIVE FORECASTING
 Executive opinions: a small group of upper level of manager may meet and
collectively develop forecast.
 it is used as long term planning and new product development
 Advantage :- bringing together the considerable knowledge and talents of these top
management people.
 Disadvantage :- generalizing the forecast by small individual and may the probability of
prevail the idea of majority.

Sale force composite:- in this case sale staff is source of information

because of it have direct contact with consumers.
disadvantage
 The sale people may be unable to distinguish between what customer would like to do and
actually will do.
 The sale people are sometimes overly influenced by the recent experience.
 CONT’D

• Consumer surveys :- the source of information for forecast is customer because it

determine the ultimate demand.
– Because of either too many customer or no way to identify all potential customer, it adbot survey
method by selecting sample and collect their opinion.
– Advantage
• survey can tap the information might not be available elsewhere.
– Disadvantage
• It need considerable amount of skill to handle the survey including collecting and analyzing the data
• It also include irrational behavior pattern

• Delphi methods
– it is opinion of manager and staff.
– It involve circulating a series of questionnaires among those individuals who possess the
knowledge and ability to contribute meaningfully.
– Responses are kept anonymous, which tends to encourage honest responses
 Steps In Delphi Method
Frist step:- establish of small group of individuals who will design the study
- This group is called Delphi committee
- They will be decide way of channel of communication , size of group required, and
participant
Send Step :- initial Questionnaire must be developed.
- It must have a tolerance level for elicits opinions
protesting the initial Questionaries’ the final questioner must be distributed for
respondents
Third Step :- the questionnaire is distributed to the participants for their initial inputs and returned to
the Delphi committee, which then compiles the result.
 Using this information, the committee develops and distributes a second questionnaire. This second
questionnaire usually asks respondents to expand on key factors or place priorities on key items
 If the ambiguity it call for feedback and re prepare the third questioners.
 And it will continue until pure common understanding is achieved for source of forecasting
•The reasons for considering a Delphi approach are:
 A group of experts can provide needed judgmental input.
 More individuals are needed than can effectively interact in a face-to- face situation, and /or the
individuals cannot be conveniently assembled in one place. Time and cost can be factors.
 It is desirable to preserve the anonymity of the participants.
 There is controlled feedback between rounds and a statistical summary of group responses.

•However, it has its own weaknesses such as:

 The questions may contain ambiguous phrasing so that panel members reach a false consensus.
 Panel membership may change, especially if the process requires a long time.
 The “experts” may not be experts.
 Preserving anonymity removes accountability and responsibility.
• Results are difficult or impossible to replicate
 Overview of quantitative method
• Qualitative methods fail in to two categories : time series model and associative/
causal models.
Time series models
 a time series is a time ordered sequence of observation taken at regular interval over
of time.
The data may be measurement of demand, earing, profits, shipment, output etc.
 It have the assumption of of future value of the series can be estimated from past
value
 Time series analysis will have two variable ; dependent and independent variable
 time is independent variable and other forecasted variable will be dependent variable .
 time serious model including the following techniques to forecast:
 Naïve forecast Exponential smoothing
 Moving average Trend projection
 A. NAÏVE FORECAST
 it is the simplest way to forecast is too assume that, the demand in the next period will be equal to
demand in the most recent period.
 However, the naïve forecast may take in to account a demand trend analysis
 Example, suppose that last week the demand was 120 units and the week before it was 108 units. So, the
forecast for next week would be 120+12= 132 units.
 Example: if the actual demand of umbrella is 60 units on Monday, the forecasted demand for Tuesday will
be 60 units.

 B. SIMPLE MOVING AVERAGES :-

 Moving average forecast uses a number of historical actual data values to generate a forecast. Moving
averages are useful if we can assume that market demand will stay fairly steady overtime.
•Mathematically, the simple moving average, (which serves as an estimate of the next period's
demand) is expressed as;

• Moving average= demand in previous n periods

N
 Where, N is the number of periods in the moving average.
CONT’D
Or
At 1  At  2  At  3  ...  At  n
SMA = Ft =
n
Where
SMA – simple moving average
Ft - Forecast for period t
At-i - Actual demand in period t-i
n - Number of periods (data points) in the moving average
 CONT’D
• There are two problem with this approach
– 1. Increasing the size of N does smooth out fluctuations better, but it makes the
method less sensitive to real changes in the data.
– 2. Moving averages cannot pick up trends very well. Because they are
averages, they will always stay within past levels and will not predict changes to
either higher or lower levels. That is, they lag the actual values
– 3. Moving averages require extensive records of past data
• Advantage
– easy to compute and easy to understand
• each new actual value becomes available for further forecast
C) WEIGHTED MOVING AVERAGE
• It involve selecting different weights for each data value and then computing a weighted mean
as the forecast
•In most cases, the most recent observation receives the most weight, and the weight decreases
for older data values. This practice makes forecasting technique more responsive to changes
because most recent periods may be more heavily weighted.
• Ft = WMA = W1At-1+W2.At-2+… +Wn.At-n
n
= A
i 1
t 1 .Wi

Where
Ft =forecast in time t
WMA = weighted moving average
W = weight
A = Actual demand value
D) EXPONENTIAL SMOOTHING
 In the previous Methods of forecasting (simple and weighted moving average),
the major drawback is the need to continually carry a large amount of historical data.
 In many applications, the most recent occurrences are more indicative of the future
than those in the more distant past
 Exponential smoothing is a sophisticated weighted moving average the most used of
all forecasting techniques.
The major reasons that exponential smoothing techniques have become so well
accepted are;
 1. Exponential models are surprisingly accurate
 2. Formulation an exponential model is relatively easy.
 3. The user can understand how the model works.
 4. There is very little computation required to use the model
 5. Tests for accuracy as to how well the model is performing are easy to compute.
CONT’D
 In the exponential smoothing method, only three pieces of data are needed to forecast the future:
 the most recent forecast
 the actual demand that occurred for the forecast period, and
 a smoothing constant alpha.
 This smoothing constant determines the level of smoothing and the speed of reaction to differentiate
between forecasts and actual occurrences
 New forecast= Old forecast + (Actual – Old forecast) or

Mathematically
Ft = Ft-1 + (At-1 - Ft-1)
Where
Ft = Forecast for period t
Ft-1 = Forecast for the previous period
  = Smoothing constant (0<  <1)
 At-1 = Actual demand for the previous period
CONT’D
• Exponential smoothing requires that the smoothing constant alpha () be given a value
between 0 and 1.
• If the real demand is stable, we would like a small alpha to lessen the effects of short term or
random changes. If the real demand is rapidly increasing or decreasing, we would like a large
alpha to try to keep up with the change.
 Associative Forecasting Techniques
Unlike, Time series forecasting associative forecasting models usually
consider several variables that are related to the quantity being predicted.
Once these related variables have been found, a statistical model is built and
used to forecast the item of interest
 For examples, the Sales of Vestel TV might be related to the company’s
advertising budget, the company’s prices, competitor’s prices and
promotional strategies, and even the nation’s economy and unemployment
rates .
The most common quantitative associative forecasting model is linear
regression analysis
 SIMPLE LINEAR REGRESSION ANALYSIS
• Regression can be defined as a functional relationship between two or more
correlated variables. It is used to predict one variable given the other. A simple
linear regression model consists of two variables that are taught to be related.
Example we can expect that
•  An extended period of rain will increase sales of umbrellas
•  The cost of a text book may depend on the number of pages
•  The yield from a land depends on soil conditions, amount and timing of rain
and fertilizer applications.
• One of these is the variable to be forecasted, and it is referred to as the
dependent variable; its value “depends” on (is a function of) the value of another
variable.
• The other variable, which will be used to “explain” or predict the value of the
dependent variable, is referred to as an independent variable
CONT’D

• The relationship is usually developed from observed data. In order to

make a forecast, the two variables must be expressed in a linear equation
of the form
• Y = a + bx
• Where
• Y= the dependent variable
• x= the independent variable (in time series analysis, X is unit of time)
• a= value of y when X=0
• b= slope of the line
• Linear regression is useful for long-term forecasting of major occurrences and aggregate
planning. The major restriction in using linear regression forecasting is, as the name implies,
that past data and future projections are assumed to fall about a straight line
• For example; the marketing manager of a shoe company has developed the following equation
to predict the demand of shoes (dependent variable) based on the amount of television
advertising (independent variable) during the holiday week:
• Y= 3.57 + 3.78x
• Where
• x= television advertising in thousands of birr in preceding week
• y= sales of shoes in thousands of birr
• Thus, the marketing manager must feel that advertising expenditure is a major factor in
determining sales of this product. According to the equation, if adverting expenditure is birr
1(thousands), sales will be about
• 3.57+ 3.78 (1) = 7.35 thousands = birr 7350
Note: It is conventional to represent values of the predicted variable on the y-axis and value of the
predictor variable on the x axis. The coefficient of a and b of the line are computed using these two
equations


b=
 XY  n xy
=
 x  nx
2 2

a=
 y  b x  
or Y  b X
n
Where n= number of paired observation Example