Considered by many to be the "bible of credit scoring," Credit Scoring and Its Applications
Expanding credit access to underserved populations (e.g., no credit history) by inferring creditworthiness from alternative data.
: Assessing existing customers to adjust credit limits, interest rates, or marketing strategies based on their repayment history and ongoing behavior. Amazon.com Key Methodologies and Tools credit scoring and its applications by l c thomas hot
As we move into an era of decentralized finance (DeFi) and on-chain credit protocols, the statistical rigor of Thomas’s framework is the only thing preventing the wild west of crypto lending from total collapse.
Reject inference is necessary when acceptance rates are low (<20%), but all methods introduce bias. The best defense is to design experiments that accept a random sample of borderline applicants to create unbiased data. Considered by many to be the "bible of
Therefore, it is now used in each of the four R's – Risk, Response, Revenue, and Retention. The University of Edinburgh
: Determining whether to grant credit to a new applicant based on their initial profile and estimated probability of default. Behavioral Scoring Reject inference is necessary when acceptance rates are
Deciding whether to grant credit to a new applicant.