ZENITH International Journal of Business Economics & Management Research

Multiple discriminant analysis is the appropriate statistical technique when the research problem involves a single categorical dependent variable and several metric independent variables. The results of discriminant analysis can assist in profiling the inter group characteristics of the subject and in assigning them to their appropriate groups.

The roles and importance of credit cards nowadays are clearly significant. The credit card users can spend future's money today. The purpose of the study is to screen credit cards consumers as high risk or low risk of Ahmedabad city. In this study, the focus is only on bank credit cards. If we take a little survey on credit cards, we can see mixed results from the people.

A small study of 35 families from Ahmedabad-Gujarat-India regarding their credit card usage is taken to set up a system to screen applicants and classify them as either ‘Low Risk’ or ‘High Risk’ (risk of default on credit card bill payment) based on information collected from questioners.

The objective of the study was to determine which variable(s) (no. of credit card used, Age of house hold head, no. of children, years of marriage, family monthly income, credit card usage frequency) are relatively better in discriminating between low and high risk applicants. And also how to classify new credit card applicants in to one of the two groups – ‘Low Risk’ OR ‘High Risk’.

The output shows that monthly credit card usage frequency is the best predictor with the coefficient of 0.839, followed by family monthly income, number of years of marriage, number of credit cards, number of children and lastly age of house hold head.

From unstandardised canonical discriminant function and means of canonical variables we can classify applicants as any discriminant score to the left of the midpoint 0.9791 leads to a classification in the low risk group. Therefore we should give this person a credit card, as he is a low risk customer. The same process is to be followed for any new applicant. If it's discriminant score is to the right of the midpoint of 0.9791, customer should be denied a credit card, as it is a high risk customer.