MotoLease LLC improved the accuracy of its internal credit forecasting model called M-Score 2.0 last month by using machine learning techniques and alternative data, according to a document shared with Powersports Finance.
The company claims the new scoring model predicts consumer creditworthiness with 66.2% accuracy, compared with 63.5% accuracy for its previous version, called M-Score 1.0, and 60% accuracy for Fico scores.
M-Score 2.0 uses machine learning — a form of artificial intelligence that allows technology to learn without manual input — to “better predict performance of new lease originations using traditional and enriched alternative data,” according to the document.
“We used machine learning techniques so that our score is not a static score, but it changes and improves based on portfolio performance,” Managing Partner Emre Ucer told Powersports Finance. “For that reason, the importance of Fico scores went down because of everything else we put into this ‘secret sauce.’”
MotoLease has been reviewing its scoring in the past year, and conducted an “extensive” study with TransUnion analyzing almost 50,000 consumer files, Ucer said.
“We came up with a very sophisticated algorithm” to underwrite subprime customers differently from prime or near-prime customers, he added. For thin-file or lower credit consumers, MotoLease reviews the consumer’s payment history, past delinquent accounts, and other payment patterns from the past 82 months.