Description: As the Valuation Manual continues to evolve and PBR standards for additional products are drafted and incorporated, it will become increasingly important for companies to understand where the requirements are consistent across products, as well as how they differ. This session will scan the life (VM-20) and annuity (VM-21, VM-22) PBR standards across the Valuation Manual to compare and contrast key requirements. Attendees will understand where common methods and techniques can be used across products and platforms. Attendees will also be better positioned to consider assumption and margin setting, asset and derivative modeling, and model efficiency techniques for all types of principles-based valuations. Additionally, this session will profile specific challenges that can arise when actuaries use models to project PBR reserves at future dates for internal planning, pricing, and projections across life and annuity businesses.

Description: Techniques that have gained some traction in general (property/casualty) insurance will be highlighted with a particular focus on the linear mixed model. This model adds random effects to the more commonly seen fixed effects (for example, as used in GLMs).

Description: In Medicaid rate development, actuaries have not historically shared models that developed the margin (i.e., underwriting gain) assumption. This is unfortunate, given the margin assumption is one of the most critical components of the rate development in ensuring financial solvency of the Medicaid managed care organizations (MCOs). In this session, we will discuss the primary components of the this assumption and will share working models that help actuaries quantify these components—including the cost of capital and risk (or contingency) margin. We expect a lively discussion as we “open the black box” and discuss these actuarial assumptions in the public for the first time.

Description: Learn about business Analytics in Insurance - Practical case studies and implementation.

Description: This session will explore the various methods of determining the credibility levels that are given to various size employer groups and Medicare plans. The background and theory of the methods that have commonly been used will be explained and then some new data-driven research in the area and possible alternatives will be discussed.

Description: You’ve built or updated a predictive model. Is it any good? How can you interpret and understand its predictions? How does it compare to alternative models? This session will look at traditional and non-traditional ways of validating predictive models, with examples from case studies.

Description: Basic feature engineering uses input data to create dummy variables, log transforms, and other single-value transformations. This talk builds upon these univariate transformations and presents several strategies for creating complex multivariate features that can enhance your model's lift. Topics such as target encoding, dimension reduction/projection, row/column, and other feature engineering topics will be covered.

Description: As of 2019, most companies have defined a model steward role. However, this role is defined differently across the industry and the people in this role face a variety of challenges. The expert panel will share background, war stories and best practices from their years of experience in the role.

Description: Principle-based reserving requires modeling assets and liabilities across a range of scenarios to assess the sufficiency of assets when determining reserves, a major shift from the current formulaic framework. These changes will require valuation actuaries to evolve their skill set by heightening their focus on creative analysis, gaining deeper understanding of complex projections and clearly communicating drivers of reserve movements to stakeholders in order to influence strategic decision-making. The panelists will discuss the way that they see the role of the valuation actuary changing in response to the change in focus in reserving requirements. They will also discuss successful characteristics along with best practices for understanding and explaining results in the new paradigm. The audience will be given the opportunity to interact through polling and facilitated discussion of their stories, experiences and observations.

Description: Discussion of machine learning applications currently being used to augment traditional actuarial modeling in the life insurance industry. This includes data science techniques for: Quantifying and explaining differences between model versions or reporting periods Building proxy models for actuarial model output using machine learning One-class classification for data integrity checks Applications of unsupervised learning (e.g., clustering and principal components analysis) to reduce model run time