VantageScore White Paper Explains VantageScore 4.0’s Use of Machine Learning
Technology expands the universe of borrowers to include more consumers with limited credit histories while maintaining regulatory compliance.
VantageScore Solutions released a new white paper that showcases the benefits of using machine learning to score more people with greater accuracy.
The white paper, “Scoring Credit Invisibles: Using machine learning techniques to score consumers with sparse credit histories,” provides a detailed description of the development process and integration of machine learning into the new VantageScore 4.0 credit score model, which is now commercially available.
VantageScore 4.0 is the first and only tri-bureau credit scoring model to incorporate machine-learning techniques to develop multidimensional attributes for consumers with limited credit histories. This development yielded significant performance increases among “dormant” consumers (those who had no update to their credit file in the prior six months). Data analysts then aligned and fully integrated these redesigned attributes and scorecards into a traditional scoring algorithm with a focus on compliance concerns.
Using machine learning techniques has led to a performance lift of as much as 16.6% for bank card originations and 12.5% on auto originations of consumers with dormant credit files. This innovation further bridges the gap between access to mainstream credit and those consumers without deep credit histories who have traditionally been frozen out of lenders’ automated underwriting systems.
“Part of our mission at VantageScore is to enhance our models so they stay fresh and relevant to the current economy and the state of consumers today,” said Sarah Davies, senior vice president, research, analytics and product development, VantageScore Solutions. “There remain millions of so-called credit invisible in the U.S. today, many of whom are creditworthy. We have a responsibility to update our models so they can accurately score more people without lowering credit risk standards and provide lenders with models that they can plug into processes with relative ease.”
Machine learning has been a part of the technology world for decades, but VantageScore 4.0 is the first and only tri-bureau credit scoring model to utilize its power. Setting it apart from conventional models that typically contain model attributes that incorporate only one or two dimensions of behavioral credit data (e.g., number of inquiries, mortgage balance, etc.), VantageScore 4.0 takes into account multiple behavioral dimensions while using innovative modeling techniques to score more consumers.
The white paper also discusses how model developers can enable FCRA-compliant model attributes that result in the use of machine learning.
To read the “Scoring Credit Invisibles: Using machine learning techniques to score consumers with sparse credit histories” white paper, visit www.vantagescore.com/machinelearningWP.
Popular Articles
Consumer FAQ: Benefits of Adding Rent and Utility Data to a Credit File
Advantage of Adding Rent and Utility Data whitepaper
Credit with a Conscience fact sheet
Driving Financial Inclusion with Data and Analytics fact sheet