Financial institutions build predictive models using Oracle R Enterprise to speed model deployment
By Mark Hornick on May 30, 2014
See the Oracle press release, Financial Institutions Leverage Metadata Driven Modeling Capability Built on the Oracle R Enterprise Platform to Accelerate Model Deployment and Streamline Governance for a description where a "unified environment for analytics data management and model lifecycle management brings the power and flexibility of the open source R statistical platform, delivered via the in-database Oracle R Enterprise engine to support open standards compliance."
Through its integration with Oracle R Enterprise, Oracle Financial Services Analytical Applications provides "productivity, management, and governance benefits to financial institutions, including the ability to:
- Centrally manage and control models in a single, enterprise model repository, allowing for consistent management and application of security and IT governance policies across enterprise assets
- Reuse models and rapidly integrate with applications by exposing models as services
- Accelerate development with seeded models and common modeling and statistical techniques available out-of-the-box
- Cut risk and speed model deployment by testing and tuning models with production data while working within a safe sandbox
- Support compliance with regulatory requirements by carrying out comprehensive stress testing, which captures the effects of adverse risk events that are not estimated by standard statistical and business models. This approach supplements the modeling process and supports compliance with the Pillar I and the Internal Capital Adequacy Assessment Process stress testing requirements of the Basel II Accord
- Improve performance by deploying and running models co-resident with data. Oracle R Enterprise engines run in database, virtually eliminating the need to move data to and from client machines, thereby reducing latency and improving security"