The Concept of Demand:-

A desire for a commodity back by the ability & willingness to

pay for it.

Factors affecting Demand:-

• Price:- Change in price & variation in demand arises

• Income Levels:- Higher income – Higher Demand
• Consumer Preference's: - Trends and Taste significantly
affect demand
The Law of Demand:-
Given by Alfred Marshall:-

The law of demand states that the quantity demanded of a good

shows an inverse relationship with the price of a good when after
factor are held constant.
Market and Elasticity of Demand :-
Market Demand:-
It can be defined as the sum of all the individual demand for a
product at price, at a point of time aka, the aggregation of
individual demand at a given price & at a point.

Elasticity of Demand:-
It is defined as the degree of responsiveness of demand for a
product to change in its determinant.
Ed = dQ/dF x F/Q
Introduction to Demand
Forecasting
Definition:
Demand forecasting is the process of predicting future customer
demand for a product or service based on historical data and market
analysis.

Importance:
Helps in inventory management, production planning, budgeting, and
minimizing costs.
Pillars of Demand forecasting:-
Current Demand:
The
v purpose of estimating current demand for the product is to
plan an appropriate level of short-term production and the price
of the product, given the market conditions…

Future Demand:
The purpose of estimating future demand is to have the
knowledge for making long-term plans for future production,
product pricing, capital investments, organising inventories for
sales promotion by advertisement, if required…
Objectives of Demand
Forecasting
 Estimate future demand.
 Optimize inventory levels.
 Improve customer satisfaction.
 Minimize costs and avoid overproduction or
stockouts.
 Example: A retail store forecasting demand
to stock the right amount of seasonal goods.
Process of Demand Forecasting
1. Specification of the Objective.
2. Determination of Time Perspective.
3. Determination and Collection of Required
Data.
4. Specifying the method of Demand
Estimation.
5. Data analysis and Derivation of
Conclusions.
Types of Demand Forecasting

Active
Passiv
e

DEMAND
FORCAST
ING

Short Extern
Term al
Long Intern
Term al
Methods of Demand Forecasting
 Survey Methods
 Consumer Survey - direct interview.
 Opinion Poll Methods.

 Statistical Method
 Trend Projection.
 Barometric Method.

 Econometric Method
 Regression Method
 Simultaneous Equation Method
Survey Method: -
In this method the company or desired person directly interview to the customer
or end-customer.
 Complete Enumeration: -
• It is used when market size is small and we can ask the question to all Consumers.
• In this method the quantities indicated by the customer are added to find the probable
Demand of customer
Dp = q1 + q2 + q3 +….+qn
where,
Dp = Probable Demand
q = demand by the Individual Households Cont…
Limitation:-
1. For very small Market.
2. Hypothetical Answer.
3. Based on own Expectation.
4. Not willing to give right answer i.e. Biased Answer.

Sample Survey Method:-

When the market size is large so we survey the potential customer as a sample
Dp = Hr / Hs (H x Ad)
Advantage:-
• Less time and Less Cost

• More Reliable

• Yearly all households and Government make budget so forecasting can be easily done.
End User Method:-
This is basically done for forecasting demand for Inputs.
• Identify the potential consumers
• Fixing Norms
• Application of Nomrs
• Collect Data according to product-wise and use-wise.

Advantages:-
• Do not give only an aggregate figure as other
• Helps in actual assumption and estimation for future.
Opinion Poll Method:-
This method is aimed to collect the opinions from those possess some
knowledge of market example sale representatives, sale executive, consultant
etc.
a) Expert Opinion b)Delphi Method c)Market Studies and Experiment

Expert Opinion:-
Firms having a good network of sales representative can be given a set of
questionnaire and ask to fill it
Limitation:-
• Based on SR skills
• Can lead to over or under estimation
• Sales representatives have narrow view of market
Delphi Method:-
The Delphi method seeks as a group of experts are asked to make
a report and submit it to group leader.

Then group leader changes and intermix the report within the
experts and ask for suggestions.

Then again report is re-distributed to the original one’s and ask

for final estimation after reading the suggestion.
Market Studies and Experiment:-
It is done to check the elasticity coefficient and variables of
demand.
First Step is to choose 3-4 city or places which similar on various
parameters

Make the Changes in Price , Branding, Advertisement, Place of

selling etc.

Then interpret and analyze the changes and variations.

Trend Projection Method:-
In this method, Demand Forecasting is done based on the
previous trends or past experiences.
It is of 3 types:-
a)Graphical Method
b)Trend Fitting
c)Box-Jenkins
Trend Graphical Method: -

Predictions
Actual
Mid
Trend Values of
Variables
Straight
line
Trend

Now the basic limitation is that Trend Line can changed according to
manager or analyzer and hence sometimes its less reliable
Trend Fitting Equation / Least Square Method:-
Fitting trends equation is a formal technique of projecting the trend in demand.
Under this method, a trend line (or curve) is fitted to the time-series sales data with
the aid of statistical techniques.
It is of two types:-
a) Liner Trends b)Exponential Trends

Linear Trend:- When a time-series data reveals a rising trend in sales, then a
Straight line trend equation of the following form is fitted.
S = a + bT
Σ S = na + bΣ T
Σ ST = aΣ T + bΣ T 2
Exponential Trend:-
When the total sale (or any dependent variable) has increased over the past
years at an increasing rate or at a constant percentage rate per time unit,
then the appropriate trend equation to be used is an exponential trend
equation of any of the following forms.
Y = aebT
log Y = log a + bT

Limitation : -

• We asume that changes in the past will persist in future too.

• Cannot be used for short term

• Does not bring dependent and independent variables.

Box-Jenkins Methods:-
When sales data of various commodities are plotted, in casse of many commodities,
it will show a seasonal or temporal variation in sales. For examples, sale of woollen
clothes will show a hump during months of winter in all the years under reference.
The sale of new year greeting cards will be particularly high in the last week of
December every year. Similarly the sale of desert coolers is very high during the
summers each year. This is called seasonal variation. Box-Jenkins technique is used
for predicting demand where time-series sales data reveals this kind of seasonal
variation.

Auto Regression Moving Average Auto-Regressive &

Model Model Moving Average Model
Auto-Regression Model:-
In a general autoregressive model, the behavior of a variable in a period is linked to
the behaviour of the variable in future periods
Yt = a 1 Yt–1 + a 2 Yt–2 + …+ an Yt– n + et
Moving Average Model:-
The moving average model estimates Yt in relation to residuals (et) of the previous
years. The general form of moving average model is given below
Yt = m + b1et – 1 + b2et – 2 …+ bp et– p + et
Autoregressive-Moving Average Model:-
After moving average model is estimated, it is combined with autoregressive model to
form the final form of the Box-Jenkins model, called autoregressive-moving average
model
Yt = a1 Yt–1 + a2Yt–2 + …+ an Yt–n + b1et–1 + b2et–2+ …+ bp et–p + et
Barometric Method:-
The barometric method of forecasting follows the method that meteorologists use in weather
forecasting. Meteorologists use the barometer to forecast weather conditions on the basis of
movements of mercury in the barometer. Following the logic of this method, many economists
use economic indicators as a barometer to forecast trends in business activities. This method was
first developed and used in the 1920s by the Harvard Economic Service. It was, however,
abandoned as it had failed to predict the Great Depression of the 1930s.12 The barometric
technique was however revived, refined and developed further in the late 1930s by the National
Bureau of Economic Research (NBER) of the US. It has since then been used often to forecast
business cycles in the US.
Leading

Lead-Lag Indicator Co-Incidental

Lagging

Diffuse Index
Econometric Method: -
Regression analysis is the most popular method of demand estimation. This method combines
economic theory and statistical techniques of estimation. Economic theory is employed to specify
the determinants of demand and to determine the nature of the relationship between the demand
for a product and its determinants. Economic theory thus helps in determining the general form of
demand function. Statistical techniques are employed to estimate the values of parameters in the
estimated equation. In regression technique of demand forecasting, one needs to estimate the
demand function for a product. Recall that in estimating a demand function, demand is a
‘dependent variable’ and the variables that determine the demand are called ‘independent’ or
‘explanatory’ variables.
Bivariate
Regression

Regression Method Multivariate

Regression

Simultaneous Equation Method

Simple or Bivariate Regression Technique: -
In simple regression technique, a single independent variable is used to estimate a statistical value
of the ‘dependent variable’, that is, the variable to be forecast. The technique is similar to trend
fitting.
Y = a + bX

Σ XiYi = Σ Xi a + bX 2i

Multi-variate Regression:-
The multi-variate regression equation is used where demand for a commodity is considered to be
a function of more than one explanatory variables.

Qx = a – bPx + cY + dPy + jA
OR

Qx = a Pbx Yc Pdy Aj
Simultaneous Equation Model:-
In contrast, the simultaneous equations model of forecasting involves several simultaneous
equations. These equations are, generally, behavioural equations, mathematical identities, and
market-clearing equations. Furthermore, regression technique assumes one-way causation, i.e.,
only the independent variables cause variations in the dependent variable, not vice versa. In
simple words, regression technique assumes that a dependent variable affects in no way the
independent variables.

Define the variable:-

i.e. Exogenous Variable or Endogenous Variable

Collect the time series Data and adjust

Form the equation

Tools and Software for Demand
Forecasting: -
Popular Tools:
- Microsoft Excel (for basic forecasting)
- SAP Integrated Business Planning
- Oracle Demantra
- SAS Demand Forecasting

Example: Analysing how SAP IBP helps large corporations

forecast demand.
Lets Start
Q&A