Reject Inference Methods

2. Simple Augmentation

The simplest way to include information on rejects is to evaluate them using the existing scorecard; rejected loan applications can be used afterwards to adjust the scorecard.

This method is called Simple Augmentation.

The first step involves developing a scorecard using the information on approved loan applications:

Using the resulting scorecard, we can evaluate the set of rejects; rejects are evaluated as "good" or "bad" based on the acceptable bad rate value.

Finally, a joint set of approved and rejected applications is used to adjust the parameters of the scorecard:

The most obvious drawback of the Simple Augmentation algorithm is the fact that rejects are evaluated using only the actual scorecard. In fact, the scorecard can cause errors, for example, due to differences in the characteristics of the credit portfolio and those in the set of rejects.

Besides, the algorithm does not take into account the probability of a loan applications being approved or rejected. Using the algorithms results in treating approval chances of approved and rejected loan applications as identical, which is undoubtedly not true.

1. What is Reject Inference

3. Fuzzy Augmentation