Demand Estimation Final

8/4/2019 Demand Estimation Final

1/21

DEMAND ESTIMATIONAND

ITS METHODS


2/21

DEMAND ESTIMATION

TECHNIQUES

QUALITATIVE

TECHNIQUES

Consumer survey

Market experiments

Consumer clinics Virtual shopping

QUANTATIVE

TECHNIQUES

Statistical techniques

Regression


3/21

CONSUMER SURVEYS

Gathering of

information

Surveys are conducted

to assess consumers

perception of various

aspects

Drawback of this

method


4/21

Qualitative techniques

MARKET EXPERIMENTATIONS

Seller of a product introduces variations and tries it

out in a representative market

High cost technique

Advantage


5/21

CONSUMER

CLINICSConsumers are asked to act

in a simulated situation

These are laboratory

experiments


6/21

A representative sample of

consumers shop in a virtual

store simulated on the

computer screen

Eliminates the high cost in

terms of time and money

Consumer reaction

recorded

VIRTUAL SHOPPING


7/21

Demandforecasting


8/21

TECHNIQUES OF DEMAND

FORECASTING

1) Experts Opinion Poll

2) Reasoned Opinion-Delphi Technique

3) Consumers Survey- Complete Enumeration

4) Consumer Survey-Sample Survey5) End-user Method of Consumers Survey

SIMPLE

SURVEYMETHODS

1) Time series analysis or trend method2) Barometric Techniques or Lead-Lag indicators

method

3) Correlation and Regression

COMPLEX

STATISTICAL

METHODS


9/21

EXPERTS OPINION POLL

Experts are requested to give

their opinion or feel about the

product

Number of such experts is large

and their experience-based

reactions are different

Also called the hunch method

Simple Survey Method


10/21

REASONED OPINION - DELPHI TECHNIQUE

Relies on a panel of independent experts

Experts answer questionnaires in two or morerounds

Process is stopped aftera pre-defined stop criterionand the mean ormedian scores of the final roundsdetermine the results
http://en.wikipedia.org/wiki/Meanhttp://en.wikipedia.org/wiki/Medianhttp://en.wikipedia.org/wiki/Meanhttp://en.wikipedia.org/wiki/Medianhttp://en.wikipedia.org/wiki/Medianhttp://en.wikipedia.org/wiki/Mean


11/21

CONSUMERS SURVEY- COMPLETE

ENUMERATION METHOD

Forecaster undertakes a complete

survey of all consumers whose demand

he intends to forecast

Principle merit of this method

Not feasible where a large number ofconsumers are involved


12/21

Select a few consuming units out

of the relevant population

Total demand of sample units isfinally blown up to generate the

total demand forecast.

Comparison to the formersurvey

Sampling error can decrease

with every increase in sample size

CONSUMER SURVEY-SAMPLE SURVEY

METHOD


13/21

END-USER METHOD OF CONSUMERS

SURVEY

Sales of a product are projected

through a survey of its end-users

A product is used for final

consumption

Demands for final consumption

and exports net of imports


14/21

Relies on historical data to predict the future

The simplest form of forecasting using trend analysis is theprojection into the future of the current value of aneconomic variable

Advantage

Limitation

Appropriate for long-run forecasts, but inappropriate forshort-run forecasts

TREND ANALYSIS

E.g. One might forecast that next year sales would be a function of salesin the existing year or alternately next year sales would be a function ofthis years sales and the change in sales between this year and last year.Or a forecaster might predict next year sales based on sketching a linethat appears to best fit the historical data


15/21

BAROMETRIC TECHNIQUES OR LEAD-

LAG INDICATORS METHOD

Use of current values of certain economic

variable called indicators to predict the futurevalues of other economic variable

leading indicator

coincident variables

lagging indicators

Ideally, changes in leading indicators

consistently precede changes in values of other

variables

EXAMPLE


16/21

REGRESSION AND CORRELATION

Use of econometric methods to determine the nature and

degree of association between/among a set of variables

Relationship between a dependent variable and one or more

independent variables

The relationship may be expressed in the form of a demand

function

The principle advantage of this method

Technique has got both explanatory and predictive value

Use not only time-series data but also cross section data

Only precaution


17/21

A SIMPLE LINEAR REGRESSION MODEL

Case where demand is stated as that between the dependent

variable and only one independent variable is a simple linear

regression model

Regression analysis attempts to fit the best possible linear

relationship between the dependent and the independentvariable

Y = a + BX + u

Y is the dependent variable (Petrol consumption)X is the independent variable (distance traveled)

a is the intercept ; b is the slope; u is the error

term.


18/21

PROBLEMS IN REGRESSION

MULTICOLLINEARITYIDENTIFICATION

PROBLEM


19/21

SIMULTANEOUS EQUATIONS

METHOD

Also known as the complete system approach or

econometric model building

It involves simultaneous considerations of all variables

Normally used in macro-level forecasting for the economy

Indeed very complicated

Principle advantage

The values of exogenous variables are easier to predict than

those of the endogenous variables


20/21

EXAMPLE

Case examines the wide

variation in the demand for call

centers.

Importance of demand

forecasting for call centers

Difficulties involved in

estimating the demand for call

centers

Various strategies that can

be adopted to overcome

uneven demand conditions


21/21

NIYAZ QURESHI 63

RAHAT ADENWALLA 73

SAAD SYED 83

SHEHNAZ SHAIKH 93

SURAJ THAKUR 103

VISHAL MANDOWARA 113

Demand Estimation Final

Documents