Day 9: Discriminant Analysis: Team B

Discriminant Analysis is a very important tool in SPSS that works in the lines of regression and helps in predicting outcome. The function involves making of a linear equation like regression, with sets of independent variables that will predict the outcome of the dependent variable.

For instance, the credit rating (CR) of a bank customer, in other words, his capacity of defaulting on a loan will depend on the following independent variables:

·         Income

·         Debt to income ratio

·         Credit card debt

·         Other debts

·         Instances of previous defaults

·         Years with current employer, etc.

The general form of equation is:

D= a + vX₁ + vX₂+ vX₃ +....

Where

 D= Discriminant function

a= constant

v= the discriminant coefficient or weight for that variable

Xn= the independent variables (like in the above bank loan example)

The aim of the statistical analysis in this type of analysis is to combine (weight) the variable scores in some way so that a single new composite variable, the discriminant score, is produced. Based on this score, one can predict how (as in the above case) liable is the next customer to make a loan default.

Applications of Discriminant Analysis in Marketing

Market Segmentation

This technique is extensively used for market segmentation by predicting a group membership and its ability to classify objects into two uniquely defined classifications. Based on the discriminant score one can predict to which group (a purchaser of the manufacturer’s brand or a competitive brand) the particular individual belongs.

The discriminant weights of each predictive variable (age, sex, income, etc) indicate the relative importance of each variable.  For instance, if age has a low discriminant weight then it is less important than the other variables.

Product research

Discriminant Analysis can be used to distinguish between heavy, medium, and light users of a product in terms of their consumption habits and lifestyles

Perception/Image research

To distinguish between groups of customers who exhibit favourable perceptions of a store or company and those who do not. Further it can be extended to predict which future customer will show a favourable response to the product displayed in the store.

Advertising research

To identify how different market segments differ in media consumption habits

Direct marketing

To identify the characteristics of consumers who will respond to a direct marketing campaign and those who will not

--

Neeraj Gandhi