Predictive Analytics Experience Studies
The SOA offers examples of experience studies that have made use of predictive analytic techniques. Most reports are in PDF format.
- 2018 Group Long-Term Disability Experience Study - New Variables and Predictive Analytics Applications
This report is a follow-up to the 2018 Group Long-Term Disability Experience Study Report published in April, 2018.
- Beyond Actual to Table: Models in Experience Studies
Traditional life experience studies show actual claims as a ratio of a reference table. Despite the issue that mixing heterogeneous segments can distort results, segments are typically aggregated to gain a larger volume of experience. This paper demonstrates an approach to addressing the issue by modeling experience, and then interpreting the model and residual errors instead of actual / table ratios alone.
- Predicting Group Long Term Disability Recovery and Mortality Rates using Tree Models
Tree models have been found to provide a good fit to group long term disability recovery and mortality rates. This research paper provides an overview of tree models, and details how they have been used for developing these rates.
- 2000-2011 Long Term Care Experience Basic Table Development
- Lapse Modeling for the Post-Level Period – A Practical Application of Predictive Modeling