Reject Inference Techniques Implemented in Credit Scoring for SAS® Enterprise MinerTM

Many business elements are used to develop credit scorecards. Reject inference, related to the issue of sample bias, is one of the key processes required to build relevant application scorecards and is vital in creating successful scorecards. Reject inference is used to assign a target class (that is, a good or bad designation) to applications that were rejected by the financial institution and to applicants who refused the financial institution’s offer. This paper discusses the technical concepts in reject inference and the methodology behind the reject inference algorithms that are available in Credit Scoring for SAS Enterprise Miner. Each reject inference algorithm is discussed in detail, and each algorithm’s impact on the final scorecard is shown. Suggestions for which algorithm to use under certain conditions are also given. OVERVIEW OF SCORECARDS Credit scorecard development is a method of modeling potential risk of credit applicants. It involves using different statistical techniques and past historical data to create a scorecard that financial institutions use to assess credit applicants in terms of risk. A scorecard model is built from a number of characteristic inputs. Each characteristic is comprised of a number of attributes. In the example scorecard shown in Figure 1, age is a characteristic and “25–33” is an attribute. Each attribute is associated with a number of scorecard points. These scorecard points are statistically assigned to differentiate risk, based on the predictive power of the variables, correlation between the variables, and business considerations. For example, in Figure 1, the credit application of a 32 year old person, who owns his own home and makes $30,000, would be accepted for credit by this institution. The total score of an applicant is the sum of the scores for each attribute present in the scorecard. Smaller scores imply a higher risk of default, and vice versa. Figure 1: Example Scorecard REJECT INFERENCE One of the main uses of scorecards in the credit industry is application scoring. When an applicant approaches a financial institution and applies for credit, the institution has to determine the likelihood of this applicant repaying the loan on time or of defaulting. Financial institutions are constantly seeking to update and improve scorecard models in an attempt to identify crucial characteristics that will help in distinguishing between good and bad risks in the future.