(Photo credit: Peripitus.)
Willis Towers Watson (Arlington, Va.) has announced the release of version 4.2 of its Radar pricing software which the company says is designed to achieve agile, sophisticated pricing and a new level of market responsiveness and analytical sophistication. The release includes a new gradient boosting machine (GBM) algorithm that delivers major performance improvement, fitting GBMs up to six times faster than the previous edition, according to Willis Towers Watson. The improvement enables GBMs to be calibrated with larger sets of data and performed more easily, saving analysts time in creating deployable models.
Radar 4.2 also allows access to data, calculations or services, such as advanced machine learning models, outside of the Willis Towers Watson pricing suite via a new external callout component. This capability gives access to a wide range of information insurers may hold externally to their Radar models, reducing the complexity of any integration while offering a flexible and extendable way to enrich their pricing solution, according to Willis Towers Watson.
Addressing Pressure to Employ New Data Sources
“This is the latest in a series of upgrades, continuing Radar’s evolution and our commitment to providing insurers with the next-generation tools needed to face new challenges presented by an industry in constant change,” comments Neil Chapman, global product leader, Pricing Products Claims and Underwriting, Willis Towers Watson. “With insurers facing continual pressure to employ new data sources and advanced analytical techniques, Radar 4.2 is a real game-changer that presents innovative features that deliver richer insights and greater flexibility, as well as improved speed to market, pricing accuracy and operational efficiency.”