Description: As any international traveler who has tried to plug in an electronic device knows, the human race is great at devising many different ways to accomplish the same task. Predictive modeling is no different. Data scientists can choose from a wide variety of languages to prepare data, fit models, and verify results. The Rosetta Stone, which showed writing composed in three languages side by side, was instrumental in translating unfamiliar Egyptian hieroglyphics into the more common Greek language. This presentation follows a similar approach by showing the same predictive model implemented in the following languages: R, Python, Spark, MatLab, and SAS. Participants will compare and contrast each of these languages and, by doing so, build a greater understanding of those languages less familiar to them.

Description: Business success in the future depends on assumptions we make today. However, we operate in a dynamic environment and cannot know with certainty what lies ahead. Addressing the challenge requires a synergy between data science, expert judgement, and decision making to spot changing drivers of key risks and respond with appropriate business action. We will be presenting development of such an "ecosystem" technically enabled by Natural Language Processing to monitor and detect signals from news streams. NLP is of increasing importance for insurance companies. Additional application examples are available in session 23.

Description: Baseline and what-if/stressed projections of cash flows, financial statements and capital requirements are critical inputs to a range of business purposes, including forecasts, risk management, liquidity management, strategic decision making, etc. Forward looking projections have historically been inefficient and performed in silos making it difficult to scale the process as CFOs and CROs ask for more. In this session we discuss challenges of the forecast process and opportunities to meet desired future capabilities. We will also discuss “what-if” analysis and ways to support production of forward looking projections for “what-if” analysis in an efficient way.

Description: The Shiny web application framework by RStudio has become a popular way for data scientists working in R to create interactive data analytics. When the developer is ready to share their application with others, internally or externally, there are a number alternatives and factors to consider. This talk presents the main Shiny application hosting alternatives and considers their advantages and disadvantages. In addition, the steps necessary to deploy an application on Shiny Server, both on-prem and in the cloud using Microsoft Azure as an example, are described.

Description: Investment risk management has become a core competency of insurance companies, and much has been discussed on this topic. This session examines the current investment risk management practices such as product line specific hedging and macro hedging, with a focus on what still remains to be addressed—whether the tradeoff of addition risk management is worthwhile.

Description: Recent Population Cause of Death Trends and Implications for Life Insurance

Description: Actuaries play pivotal roles in insurance companies, consulting firms, and throughout the industry. With the ongoing accounting and regulatory changes, process and system modernization, and the rapidly changing environment, actuaries frequently find themselves working in multidisciplinary projects with non-actuarial teams. In order to take on broader leadership roles, actuaries are often times expected to partner with and/or lead non-actuarial personnel. How can actuaries work most effectively with non-actuaries? A panel of industry leaders across different disciplines will share their perspectives on actuaries working with and building effective relationships with non-actuarial colleagues. The panelists will cover how actuaries add value in a non-actuarial world, what aspects actuaries can focus on to be more effective across disciplines, and incorporate success stories as well as lessons learned.

Description: The NAIC has adopted revised VM-21 and AG43 for variable annuities, effective Jan. 1, 2020. The panel of presenters will provide background and an overview of the revisions, discuss detailed aspects of the new CTE Amount Standard Projection and prescribed assumptions, as well as discuss implementation considerations and the implications to hedging. In addition, the panelists will share progress on VM-23, principle-based reserves for non-variable annuities.

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

Description: Presentations from members of: Haven Life; Ladder Life; Life by Spot