Scribd
Upload
53 views8 pages

Chp4 Demand Estimation

1) There are several approaches to estimating demand, including consumer surveys, observational research, consumer clinics, market experiments, and virtual shopping simulations. 2) Regression analysis can be used to analyze the relationship between a dependent variable like sales revenue (Y) and an independent variable like advertising expenditures (X). 3) The regression line is estimated to minimize the vertical distance between the data points and the line, resulting in coefficients that best predict the dependent variable based on the independent variable.

Uploaded by

abhishekhrc
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
53 views8 pages

Chp4 Demand Estimation

1) There are several approaches to estimating demand, including consumer surveys, observational research, consumer clinics, market experiments, and virtual shopping simulations. 2) Regression analysis can be used to analyze the relationship between a dependent variable like sales revenue (Y) and an independent variable like advertising expenditures (X). 3) The regression line is estimated to minimize the vertical distance between the data points and the line, resulting in coefficients that best predict the dependent variable based on the independent variable.

Uploaded by

abhishekhrc
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 8

Demand Estimation

Approaches to demand estimation


• Consumer surveys: record the response of a sample of
consumers for changes in its and related goods’ prices,
income, credit incentives, advertising through
questionnaires.
• Observational research: Watch consumers buying in
response to the change factors above through scanners
and people meters.
• Consumer clinics: participants are given money in a
laboratory experiment set up to shop in simulated store
which are affected by changes in the factors.
Demand Estimation(contd)
• Market Experiments: By selecting several markets with
similar socio-economic environments, collect data on the
response of buyers for changes in the influence factors
including demographics in which every market faces a
different influence factor.
• Virtual Shopping: representative sample of customers
shop in a virtual store simulated on the computer screen.
During course of time various influence factors are
changed to observe the behaviour of virtual shoppers.
Introduction to Regression Analysis
• Example: Sales revenue (Y) depends on
Advertising exp. (X) (unit: mills of Rs)

Yr 1 2 3 4 5 6 7 8 9 10

X 10 9 11 12 11 12 13 13 14 15

Y 44 40 42 46 48 52 54 58 56 60
Regression (contd)
Regression (contd)
• Objective: draw the line of best fit through the
scatter points of Y and X.
• Line is: Ŷ â  b̂X
t t
• Points above and below the line are written:

Yt  â  b̂X t  e t
Regression (contd)
• Principle: draw the line in such a way that the
distance between the line and the points is
minimized (best is to minimize the square). Minimize
• n 2 n 2
 et   (Yt â  b̂X t )
t 1 1
• Minimization occurs when
 (X t  X)(Yt  Ŷ)
b̂ 
 2
 (X
t
X )
â  Y  b̂X
Regression Equation
• Estimated equation is:
• Ŷ  7.6  3.53X
t t
• Standard error (deviations) of b:

 2
 t(Y Ŷ
t
)
S 
b̂ (n  k) (X  X) 2
t
Regression
Hypothesis testing:
• Involves examining whether an estimated coefficient (such as b̂ )
is statistically different from zero. If different from zero, then the
variable associated affects the dependent variable.
• Estimated coefficient is considered to be different from zero (i.e.
significant) with better evidence only if its t-value falls on the
extreme 5% of its tail (or check the p-value in the last column of
estimation: if p-value is smaller than 0.05, then the coefficient can
be taken different from zero in statistical sense.
• Normally statisticians take p-value <0.1 (as weak evidence), p<0.05
(better evidence) and p<0.01 (strong evidence)
• R2 is a measure of goodness of fit of the regression. It is also the
square of the correlation between Y and Ŷ
• Hypothesis test needed to show whether meaningful relation exists.

You might also like