CHAPTER TWO

DEMAND ANALYSIS AND DEMAND FORECASTING

CONCEPT OF DEMAND AND ITS ELASTICITY:
• Meaning and types of demand
• Determinants of demand
• Demand Function and its Types
• Concept of elasticity of demand
• Price elasticity of demand
• Income elasticity of demand
• Cross elasticity of demand
• Promotional elasticity of demand

2
 USES OF PRICE ELASTICITY OF DEMAND IN
BUSINESS DECISION MAKING
• Product pricing
• Pricing of joint products
• Demand forecasting
• Pricing of inputs
• Financial planning
• Determination of marketing/promotional strategies
• Useful to understand the role of trade unionists etc.
Note: price elasticity of demand is also helpful in international trade and
determination of government policies as following:
3
CONTD..
• Useful in International Trade :
– Determination of terms of trade
– Determination of tariff policy
– Determination of policy of devaluation
• Useful in formulation of government policies:
– Tax policy
– Policy of granting protection
– Explaining the Paradox of Poverty

4
 USES OF INCOME ELASTICITY OF DEMAND IN
BUSINESS DECISION MAKING

– Determination of price of products

– Demand Forecasting
– Long term business planning.
– Determination of Market Strategy.
– Housing and other Development Strategies.
– Classification of Goods and market

5
 USES OF CROSS ELASTICITY OF DEMAND
IN BUSINESS DECISION MAKING

• Demand forecasting
• Formulation of business policy
• Classification of goods
• Classification of markets
• Determination of pricing strategies
• Determination of marketing/promotional strategies etc.

6
 USES OF PROMOTIONAL ELASTICITY OF
DEMAND IN BUSINESS DECISION MAKING

• Determination of marketing strategies.

• Selection of appropriate advertisement scheme.
• Understanding the competitiveness of the market.
• Judge the need of advertisement and promotions.
• Understand the strategy for market domination.
• Make customer retention and customer attraction advertisement plans and
policies etc.

7
 RELATIONSHIP BETWEEN PRICE ELASTICITY
OF DEMAND AND REVENUE
M=

• Derivation of relationship
 e 1
MR  AR  
 e 
• At e > 1, MR is positive and TR is increasing.
• At e < 1, MR is negative and TR is decreasing.
• At e = 1, MR is equal to zero and TR is maximum and constant.

8
CONTD…
e=1

e>1
e<1

e>1

e=1 e<1

9
 CONCEPT AND SIGNIFICANCE OF
DEMAND FORECASTING
• Meaning of Demand Forecasting:
– Prediction of future demand of the product on the basis of current state of relationship between the
determinants of demand.
– Forecasting aims to reduce uncertainty about tomorrow, so that effective decision can be made today
by providing predictions of future values of variables from past and present information

10
SIGNIFICANCE OF DEMAND FORECASTING
• Production planning:
• Sales planning:
• Inventory planning:
• Profit planning:
• Production Stability maintenance:
• Growth and long-term investment programing:
• Economic planning and policy making:

11
PURPOSES OF DEMAND FORECASTING
• Short-term Forecasting (Problem Solving)
(i) Evolving a suitable production and sales policy:
(ii) Helping purchase planning to reduce the cost of production:
(iii) Determining short-term financial requirements:
(iv) Determining appropriate pricing policies:
(v) Fixing the sales target:
• Medium and Long-term Forecasting (Planning/Strategy formulation)
(i) Planning the establishment of new unit or expanding the existing unit
(ii) Long-term financing planning
(iii) Human resources planning etc.

12
 STEPS IN DEMAND FORECASTING
1. Identify and clearly state the objectives of the forecasting
2. Select appropriate methods of forecasting
3. Ascertain the determinants of demand for the particular product or product group
4. Define demand function
5. Collection of Data
6.Presentation and Analysis of Data
7.Interpretation of results

13
CRITERIA FOR GOOD FORECASTING
• Accuracy
• Simplicity
• Durability
• Flexibility
• Economy
• Availability

14
TECHNIQUES OF DEMAND FORECASTING

15
SURVEY METHODS
• used to make short-term forecasts.
• surveys are conducted to collect information about consumers' intensions and their future
purchase plan.
• Basically, new products require the use of survey method only because of absence of any
historical data.
• includes: i) Consumer's survey method;
ii) Opinion survey of market experts and sales representatives; and
iii) Market studies and experiments.

16
CONSUMER'S SURVEY METHOD:
• direct interview of the potential consumers.
• ask them what quantity of the product they would be willing to buy at different prices over a
given period, say, one year.
• survey can be conducted by simply stopping and questioning people at the shopping centers or
any other places.
• carefully constructed questionnaires and trained interviewers is necessary.
• 3 alternative ways :
 Complete enumeration method;
 Sample Survey method; or
 End-use method

17
 OPINION SURVEY OF MARKET EXPERTS AND
SALES REPRESENTATIVES
• supposed to possess knowledge of the market, e.g., sales representatives, sales
executives, professional marketing experts, and consultants. Opinion survey
method includes.
a.Expert-opinion method;
– aims at collecting opinion of those who are supposed to have knowledge of the market.
– They are supposed to know about future purchase plans of their customers, their reaction to
the market changes and the demand for competing products.
– The estimate of demand thus obtained from different regions are added up together to get the
overall probable demand for a product.

18
CONTD…
b. Delphi method
– developed by Olaf Helmer, Dalkey and Gordon in 1940.
– In business management areas, its relevance is in human resource planning, demand estimation.
– similar to market opinion method.
– Panel of experts selection- view collection- revise- final forecasting

19
MARKET EXPERIMENT METHOD
• conducted in the actual market place.
• There are many ways of performing market experiments.
• One method is to select several markets with similar socio-economic characteristics
(population, income level, cultural and social background, choice and preferences).
• Market Experiment is conducted by changing commodity price in some markets or stores,
packaging in other markets or stores, and the amount and type of promotion in still other
markets or stores, the record the purchases of consumers in the different market.
• Market experiments are helpful to a firm in determining its best pricing strategy, promotional
campaigns, and product qualities.
• It is useful in the process of introducing of products for which no other data exist. And, it is
also useful in verifying the results of other statistical technique.

20
LABORATORY EXPERIMENT
• The participants are given a sum of money and asked to spend it in a stimulated store.
• Reaction of participants regarding changes in the commodity prices, product packaging,
displays, price of competing products, and other factors influencing demand will be noted.
• Participants are selected so as to closely represent the socio-economic characteristics of the
market of interest.
• The experiment reveals the consumers' responsiveness to changes made in prices packages
etc.
• laboratory experiments are more realistic than consumer survey in that it reflects actual
consumer behaviour.

21
STATISTICAL METHODS OF
DEMAND FORECASTING:

22
• Statistical methods are considered to be superior technique of
forecasting for the following reasons:
a.Method of estimation is scientific;
b. Estimation is based on the theoretical relationship between the dependent
and independent variables;
c.Estimations are relatively more reliable; and
d. Estimation involves lower amount of cost.

23
TIME SERIES ANALYSIS
• Classical method of business forecasting.
• Statistical data presented in chronological order is called time series data.
• Also known as trend projection, extrapolation or lost horse method.
• Mathematically, time series is defined as:
Y = f(t)
Where, Y = dépendent variable
t = indépendant variable
• Objective of time series are
- to study the past behavior of the data; and
- to forecast the future behavior;
• Components of Time Series
i) Secular Trend (S) ii) Seasonal variation (V) iii) Cyclical variation (C)
iv) Random or irregular fluctuation (R)
24
 TWO MODELS ARE USED TO ANALYZE THE FLUCTUATIONS OR VARIATIONS
IN TIME SERIES DATA, WHICH ARE:
a) Additive Model :
•This model analyzes the fluctuations of time series data by adding the probable fluctuation in
above stated four causes of fluctuations.
Mathematically,
Y=S+V+C+R
Where,
Y = dependent variable
S = secular trend
V = seasonal variation
C = cyclical variation
R = random or irregular fluctuation

25
CONTD…
(b) Multiplicative Model
•This model analyze the fluctuations of time series data by multiplying the probable fluctuations in
causes of fluctuations
Mathematically,
Y=S×V×C×R
Where,
Y = dependent variable
S = secular trend
V = seasonal variation
C = cyclical variation
R = random or irregular fluctuation
Given,
Y=C×V×C×R
Taking log on both sides
log Y = log C + log V + log C + log R
26
MOVING-AVERAGE METHOD:
• Simple device of reducing fluctuations and obtaining trend values with a fair degree of accuracy.
• the forecasted value of a time series in a period (month, quarter, year, etc.) is equal to the
average value of the time series in a number of previous periods.
• The forecasted value of the time series for the next period is given by the average value of the
time series in the previous three periods, under three period moving average.
• The greater the number of periods used in the moving average, the greater is the smoothing
effect because each new observation receives less weight.
(Numerical Example Needed)

27
REGRESSION ANALYSIS:
• most popular and most useful tool of determining the strength of relationship between
dependent and independent variables.
• The value of dependent variable is determined on the basis of independent variables.
• Regression line can be used to forecast the value of dependent variable which represents the
mean value of independent and dependent variable relationship.
• In case of demand forecasting, demand of a product is dependent variable and its value is
forecasted with the help of its determinants (independent variables).
• Two types of regression:
a) Simple Regression Analysis
b) Multiple Regression:

28
 SIMPLE REGRESSION ANALYSIS
• only one independent variable or demand of a product depends upon only one determinant,
• simple regression is used to forecast the value of single variable demand function.
• The least square equation for simple regression could be shown as:
Y=a+bx
Where,
'a' is intercept (constant variable)
'b' is marginal change;
'Y' is dependent variable; and
'x' is independent variable
• In the given least square equation, the value of regression coefficients a and b can be derived by solving
following two least square equations or normal equations:
Y = na + bx
xY = ax + bx2
• By solving these two equations, we get the values of a and b. When we substitute these values in
regression equation, it helps to forecast the value of dependent variable with given independent variable.

(Numerical Example Needed) 29

MULTIPLE REGRESSION:
• If the value of quantity demanded of a commodity depends on two or more determinants of demand,
we can use multiple regression equation to forecast.
Multiple regression equation can be presented as,
Y = a + b1 x1+ b2 x2
Where,
‘a’ is intercept
‘Y’ is dependent variable.
x1, x2 are independent variables.
b1, b2 are regression coefficients.
• Similarly as simple regression equation, the value of regression coefficients in multiple regression
equation can be obtained by solving the following normal equations.
Y = na + b1  x1+ b2 x2
Y x1 = a x1 + b1  x12+ b2  x1 x2
Y x2 = a x2 + b1  x1x2+ b2 x22
• The value of a, b1, b2 can be obtained by solving above stated normal equations and value of dependent
variable is forecasted by using the regression equation.
30
BAROMETRIC FORECASTING METHOD
• This method was first developed and used in the 1920s by the Harvard Economic Survey.
• revived and developed by the National Bureau of Economic Research (NBER) and the
Conference Board.
• Barometric forecasting follows the method of meteorologist use in weather forecasting.
• They use barometric to forecast weather condition on the basis of movements of mercury in
the barometer.
• The barometric technique is based on the idea that the future can be predicted from certain
happenings in the present.
• Mainly, following three types of time series/indicators could be observed in barometric
techniques of forecasting.
(i) Leading Indicators;
(ii) Coincidental Indicators; and
(iii) Lagging Indicators
31
LEADING INDICATORS

• if there is a consistent change in one series before the change in other series, that is called
leading indicator.
• The variable that moves downward before peak and moves upward before trough are called
leading indicator.

32
COINCIDENT INDICATORS:
• If two series of data frequently increase or decrease at the same time, one series
may be regarded as a coincident indicator of the other.
• For example, consumption expenditure increases due to increase in income of
the people.

33
 LAGGING INDICATORS:
• sometime series move in step or coincide with movements in general economic activities and
are therefore called coincident indicators.
• Still other follow or lag movement in economic activity and are called lagging indicators.
• For example, the bank rate is the leading indicator, the rate of interest charged by commercial
bank is coincident indicator and the rate of interest charged by the money lender is a lagging
indicator.

34
 LIMITATIONS OF DEMAND FORECASTING:
• Accuracy
• Choice of Technique
• Availability of Data
• Statistical Error
• Change in Determinants
• Study of Consumer’s Psychology
• Lack of experienced experts for forecasting
• Calculation of least possible cost
• Determination of relationship among the determinants of demand etc.

35