has introduced Analytic Dataset, a new analytic tool that provides borrower-level data in an anonymous and non-aggregated format. The dataset provides key information for researchers and modelers such as credit risk scores, geography, debt balances and delinquency status at the loan level for all types of consumer loan obligations and asset classes.
With this new solution, investors and other market participants have the ability to better model delinquency, default, loss severity and prepayment, as well as the ability to more accurately value securities and understand broader consumer credit trends. Asset-Backed Securities (ABS) and Mortgage-Backed Securities (MBS) investors, issuers, traders, and ratings agencies researchers can use the tool to analyze and model consumer payment performance across a variety of asset classes such as auto, credit card, mortgage and unsecured personal loans. Also, better modeling may give investors better predictive power to price risk and thus finance consumer debt at the best possible rates.
"This data is one of the most important advances in consumer modeling and analytics," said Professor Tomasz Piskorski, Columbia Business School.
Analytic Dataset is created from an unbiased 10 percent statistical sample of the U.S. credit population across all geographic boundaries, with historic data starting in 2005. It provides insights into the credit health and payment performance of U.S. consumers over time and across various economic cycles.
"When businesses or government entities are able to apply segmentation and perform analytics by credit quality or asset class, they can better determine important factors such as how consumers prioritize payments and the impact of behaviors of given loan types on other forms of credit," said Geoffrey Hickman, Managing Director of Government Credit and Capital Markets at Equifax. "This in turn gives them the ability to drive a deeper level of understanding and improve modeling efforts."