In the 2nd edition of Credit Scoring and Its Applications (published 2017, updated in subsequent papers), Thomas outlines the next decade:
Thomas’s treatment of is particularly notable. This technique, popularized by the Fair Isaac Corporation (FICO), involves finding a linear combination of variables that best separates "good" payers from "bad" payers. Thomas explains that while this method assumes multivariate normality (a strict statistical condition rarely met perfectly in the real world), its robustness makes it a standard industry tool. Credit Scoring And Its Applications By L C Thomas
Thomas begins with the basics. The simplest approach is the linear probability model, where the probability of default is modeled as a linear function of the borrower's characteristics. While intuitive, Thomas highlights its fatal flaw: it can predict probabilities less than 0 or greater than 1, which is mathematically impossible in the real world. In the 2nd edition of Credit Scoring and
In the fluorescent-lit archives of a fading London bank, an aging risk analyst named Miriam stumbled upon a forgotten first edition: Credit Scoring and Its Applications by L. C. Thomas. The book’s spine was cracked, its margins filled with a previous owner’s frantic pencil scratches. Miriam, who had spent thirty years manually approving small business loans, felt a strange pull. Thomas begins with the basics
(originally published in 2002) is often referred to as the "bible" of credit scoring. It serves as a comprehensive guide for statisticians and risk managers on how to build, use, and monitor mathematical models to make intelligent lending decisions. Core Objectives of the Book
Years later, retiring, Miriam placed that worn book into the hands of a young intern. “Remember,” she said, “Thomas taught us how to predict the future. But we decide which future to build.”