Hitachi Capital (UK) PLC partners with Jaywing to improve application credit scores through AI technology

  • Archetype proves success using Hitachi Capital UK’s existing data—providing up to 11% uplift in performance  

Credit risk analytics expert, Jaywing today announces a new partnership with Hitachi Capital Consumer Finance.

The Consumer Finance division has appointed Jaywing to explore the potential to significantly improve its existing application credit scores using Jaywing’s AI modelling technology, Archetype.

Hitachi Capital Consumer Finance wanted to understand the uplifts that may be possible through using a deep learning approach. Jaywing processed data from Hitachi’s recent scorecard development using Archetype to replace the traditional modelling steps. Using the software, Jaywing was able to apply appropriate constraints which reflected how Hitachi’s modellers would apply common sense rules to the inputs. These rules meant that the system would only generate model outputs adhering to Hitachi’s expectations, making models explainable to regulators and customers alike – a key differentiator compared to other Neural Network-based approaches.

Save Money & Time. Join Our Newsletter For Free. 
. Event Invites          . News      . Videos              . eMagazines 
. Analysis & Opinion     . Exclusive Reports     . Interviews
Submit

All emails include an unsubscribe link. You may opt-out at any time. See our privacy policy.
 

Using exactly the same data, Archetype demonstrated an impressive uplift of 7.2% compared to an optimum linear regression model, showing that the Archetype model had the potential to predict and prevent more bad debt or to increase the number of customers taken on without increasing bad debt levels.  On a smaller sub-prime portfolio, an uplift of 11% was seen, from a lower baseline.

Archetype is currently the only commercially available deep learning-based scoring product which solves the ‘black box’ problem around explainability in the creation of models using AI. The software replaces the role of traditional modelling tools and, unlike other approaches, offers a mathematical guarantee that its output will adhere to behaviours specified by the user for each individual variable.

Nick Gibbs, Head of Commercial & Strategy at Hitachi Capital Consumer Finance, said: “We were very interested in the role that AI could play in transforming aspects of our business’s operation. Not only did Jaywing promise uplifts through the Archetype software, they delivered them too. Archetype solves the black box problem in credit risk, and has given us food for thought in how we approach our modelling activity. I look forward to working with the Jaywing team in the future as we continue to explore the use of AI in our business.”

Martin Smith, Jaywing’s Head of Product Development said: “AI is now ready for full adoption within the credit scoring sphere due to the removal of barriers surrounding the transparency of models produced. Archetype overcomes these hurdles and, as our project with Hitachi shows, is able to generate huge uplifts. We’re looking forward to working closely with Hitachi Capital and revealing the true power of AI in the form of Archetype.”

Jaywing has over 18 years’ experience in credit risk analytics, helping lenders such as Nationwide, RBS and Coventry Building Society with scoring and modelling, IFRS 9, IRB, stress testing, ICAAP and credit grading requirements.

To find out more visit risk.jaywing.com 

Save Money & Time. Join Our Newsletter For Free. 
. Event Invites          . News      . Videos              . eMagazines 
. Analysis & Opinion     . Exclusive Reports     . Interviews
Submit

All emails include an unsubscribe link. You may opt-out at any time. See our privacy policy.
 
Close
Save Money & Time. Join Our Newsletter For Free. 
. Event Invites          . News      . Videos              . eMagazines 
. Analysis & Opinion     . Exclusive Reports     . Interviews
Submit

All emails include an unsubscribe link. You may opt-out at any time. See our privacy policy.
 
Close