Below are a number of examples of experience studies and research reports that have made use of predictive analytic techniques.
Predictive Analytics Research
Machine-Learning Methods for Insurance Applications - A Survey
The Society of Actuaries is pleased to make available a research report that provides a literature survey of methodologies applying machine learning to insurance claim modeling.
Presentation: Actuarial Risk Analysis using Predictive Analytics, Segmentation and Decomposition Techniques
This presentation highlights the use of analytics for accelerated underwriting in the U.S. Life Insurance industry.
Accuracy of Claims-Based Risk Scoring Models
Material that presents the results of a study comparing health risk scoring models. The study builds on prior SOA studies with a similar objective.
Predictive Modeling Research
Dale Hall discusses actuarial research involving predictive models and analytics.
2000-2011 Long Term Care Experience Basic Table Development
Long Term Care basic experience rate tables for claim incidence, claim termination, and claim utilization. Three multiplicative models have been developed, one for each of three tables that will calculate rates that vary by a number of product and policyholder attributes.
SOA Research Institute on GitHub
Visit the SOA Research Institute’s GitHub overview page to access repositories for code from research projects. You also can post code and make comments. To publish material not related to research, please contact David Schraub at DSchraub@soa.org. The SOA’s general GitHub page is also available.