Help Protect Your Business Proactively With Predictive Risk
There is little question that analytics has a wide variety of useful applications for the financial industry today, including improving the accuracy of predictive risk models. This helps financial services professionals better understand which customers have the highest propensity for defaulting on credit products as well as helps them make decisions about marketing campaigns, target markets and credit limits.
For one financial services firm, their ability to pinpoint customers who were most likely to default on their loans within a three- and six-month time frame was assisted through the use of a new model based on advanced analytics and historical, time-series data.
Testing a new model
A new model was built combining Equifax Advanced Decisioning Attributes, which comprises more than 500 foundational consumer credit attributes, with Equifax Dimensions™, which provides detailed information on past credit tradeline activity in attribute or raw data form. When tested against its previous predictive risk model, the financial services firm found that the new model improved the accuracy of identifying customers most likely to default on loan payments by an amazing 29 percent. Of course, results may vary for other tests.
How the test worked
The firm conducted a comparative analysis of how accurate their predictive risk models were by evaluating the Kolmogorov-Smirnov (KS) points of their existing modeling tool against the new model, built from Equifax Advanced Decisioning Attributes and Equifax Dimensions. KS points are calculated from a formula that measured the predictability of consumers to default (not pay) on their loans in 3 month and 6 month time periods. The KS point statistic represents a model’s ability to differentiate “good” credit risk from “bad.” The higher the KS value, the better the model performs at both separating the two groups and accurately identifying predictive risk in a population.
Refocused marketing efforts
Identifying customers who have the highest likelihood of defaulting on payments is a valuable tool for decision-makers. Marketing efforts — and dollars — can be directed to other customer segments that are more likely to be profitable, instead of being aimed at this customer segment. This helps save time, reduce the cost of marketing and may increase profits for financial services.
Better predictive risk for credit account management
Knowing which customers have the highest default risk before you even extend credit to them means that account management decisions may be made proactively instead of reactively to poor account performance. Equifax Dimensions™ offers analytics tools based upon actual tradeline data of consumers and helps enhance the accuracy of predictive risk models.
As technology improves, so too does the ease and speed of analyzing the vast amounts of data produced by financial transactions. This means that the results of models such as this one, based on the latest analytics tools, are likely to continuously improve, helping to save more time and money for any businesses that choose to harness their power.
Equifax Dimensions is a product that must be used in accordance with the Fair Credit Reporting Act (FCRA).
Image source: VINCENZO PINTO/AFP/Getty Images