Application of Discriminant Analysis
Application of Discriminant Analysis to the Selection of Applicants, Discriminative analysis is a statistical model such can be used to accept or refuse a prospective credit customer. The discriminate analysis is as same to regression analysis however it assumed as the observations come from two different universal sets or in credit analysis, the good and bad customers. To demonstrate let us suppose that two factors are significant in evaluating a credit applicant the quick ratio and total worth to total assets ratio.
The discriminate function will be of the form as.
f_{t }= a_{1}(X_{1}) + a_{2}(X_{2})
Whereas: X_{1} is quick ratio
X_{2} is the network to total assets
a_{1} and a_{2} are parameters
The parameters can be computed with the employ of the following equations as:
a1 = (Szz dx - Sxzdz)/Sxx Sxx - Sxz²
a2 = (Szz dx - Sxzdz)/Szz Sxx - Sxz²
Whereas: Sxx represents the variances of X_{1}
Szz represents the variances of X_{2}
Sxz is the covariance of variables of X_{1} and X_{2}
d_{x} is the difference between the average of X_{1}'s bad accounts and X_{2}'s good accounts
d_{z} represents the difference between the average of X's bad accounts and X's good accounts.
The next step is to determine the minimum cut-off value of the function below at which credit will not be given. This value is referred to as the discriminate value and is denoted by f*.
Once the discriminate function has been developed it can then be used to analyze credit applicants. The important assumption here is that new credit applicants will have the same characteristics as the ones used to develop the mode.
More than two variables can be utilized to determine the discriminate function. In that a case the discriminate function will be of the form of.
f_{t} = a_{1}x_{1} + a_{2}x_{2} + ... + a_{n}x_{n}