Day 9: Team C

Discriminant function analysis is used to determine which continuous variables discriminate between two or more naturally occurring groups. It is a very important tool in SPSS that 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.

The characteristics of discriminate analysis is as follows:

•  Outcome variable or the dependent variable will be a category variable and will assume two values unlike regression.
•   It is a predictive function – can be used to predict whether a variable will belong to a particular group.
•   To do discriminate analysis we must know past data.
• All the variables used to make the discriminate equation are normally distributed.

Discriminant function analysis is broken into a 2-step process: (1) testing significance of a set of discriminant functions, and; (2) classification. The independent variables are the predictors and the dependent variables are the groups. In the example of the bank loan taken in class, the capacity of a bank customer defaulting on a particular loan will account into the following independent variables:

• Income
• Debt to income ratio
• Credit card debt
• Other debts
•  Instances of previous defaults
•  Years with current employer

Now, the general form of the equation is:

D = b + mX₁ + mX₂ + mX₃

where;

D = Discriminant function

b = constant

m = the Discriminant coefficient

X_n = the independent variables

The objective here is to combine the variable scores in some way so as to generate the Discriminant score. Based on the score one can predict whether any future customers are like to default or not (as in the above example).

Trilochan Pariyar