Project Ellipse

Project Ellipse explores how technology solutions could enable supervision to be more forward-looking, insights-based and data-driven using an integrated regulatory data and analytics platform. Importantly, the Ellipse prototype combines both structured and unstructured sources of data that are relevant to current events in real time. Advanced analytics are then applied to those integrated data sources to provide supervisors with early warning indicators, analytics and prudential metrics.

Project Ellipse was undertaken in two phases. In Phase 1, the project investigated how data-driven supervision could be enabled by machine-executable digital reporting, using a cross-border common data model. Phase 1 illustrated the possibilities and the efficiencies that could be gained if machine-executable reporting using common data models were to be adopted. This could also increase the volume of granular data available to supervisors, as needed to enable the use of advanced analytics. You can view our Phase 1 data model, a demonstration and technical documentation.

In Phase 2, the project took existing large exposures regulatory data and integrated these with unstructured data. Advanced analytics such as machine learning and natural language processing were applied to these data sources to make risk correlations and to analyse sentiment, alerting supervisors in real time of issues that might need further investigation. Network analytics were also used to demonstrate how exposures could be mapped, indicating possible systemic risks to the banking system.

The BIS Innovation Hub's Project Ellipse is a prototype that authorities can test in their own environments, and which may help them to explore new solutions. It also presents an opportunity for the global regulatory community to further consider, explore and collaborate on common solutions to future-proof the data and analytical capabilities of supervisors.