Description: The presenters will discuss predictive analytics as applied to recent group life and disability experience studies and other applications.

Description: Model design is an important part of being an actuary, and the SOA highlights the importance of the actuarial control cycle of model design, evaluation and monitoring. What tools are at our disposal for these steps? When should we be concerned or excited about model performance? Can we, and should we, use machine learning tools to assist with model improvement or development?

Description: This paper develops a formal framework to model the behavior of individual and small group health insurance policies in a competitive market. In this framework, anti-selective lapse is not assumed, but arises as a result of rational consumers seeking the best value in a competitive market where the premium rates are restricted by law. The authors obtain a model that addresses a number of fundamental questions which have not been fully answered previously, including the effect of price-induced anti-selective lapse on the mix of insureds in a block and the optimal renewal price to maximize the aggregate profit of a block.

Description: Session 1D: Meeting the Mark - Predicting MA Stars Outcomes with Monte Carlo Simulations

Description: Gather with members and non-members in the China actuarial community to discuss recent topics and related issues in areas such as macroeconomic sharing, pension, investment, product strategy, and more.

Description: This report is presented in the form of a Tableau dashboard and is the sixth public release from this report series and contains the results of an excess mortality analysis for the fourth quarter of 2021. Data from 30 companies representing approximately 69% of the industry’s face amount inforce were included in the analyses in this report. A total of 3.2 million death claims from individual life policies from 2015 through December 31, 2021 make up the basis of the analyses. The report also compares the experience under COVID-19 of the insured population versus the general population.

Description: In this presentation, actuaries and data scientists from Cleveland Clinic and Milliman discuss the development of a sepsis prediction model using machine learning. Presented with a unique challenge of handling massive amounts of data for near real-time risk scoring, the team took their model development to the cloud and delivered a product in less than a week through a unique virtual engagement.

Description: Much of the work performed by disability insurance actuaries depends on the availability of accurate and detailed data, which is not always as simple as it sounds. Explore a variety of data challenges faced by disability actuaries, along with potential solutions. Potential topics include finding relevant data, the impact of poor data on valuation work, the impact of new privacy rules, and the potential bias in the data used to develop predictive models.