Description: In this session, we will delve into the powerful capabilities of two essential algorithms in the field of data science and machine learning: Generalized Additive Models (GAM) and CatBoost. These algorithms can play a crucial role in conducting experience studies for insurance.

Description: The importance of storytelling in financial communication and its role in conveying complex financial reporting results and insights to diverse stakeholders cannot be denied.

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.

Description: Human longevity is one of the great success stories of the past century, and there is growing consensus that there are further gains to come. What will these future gains look like, where will they come from and how likely are they to happen. Furthermore, how do we as actuaries deal with the uncertainty when trying to predict future long-term mortality improvements more than 20 years into the future. This is one of the most important assumptions on the balance sheet that insurers and reinsurers make, so it is important to have a strong perspective on it. Our presentation will give attendees a view into some of the core components of how to assess long-term mortality improvements by looking at the cyclical nature across different countries of life expectancy. It will also deep dive into the different drivers of life expectancy over the past cycle driven largely by improved diagnosis and treatment of cardiovascular disease (CVD) and cessation of smoking, which ended in about 2010. It will then discuss how population mortality improvement has slowed and plateaued in many major countries.

Description: How might retirement income strategies be designed, and their suitability assessed, when retirees may differ substantially in their objectives, preferences and personal circumstances? This webcast provides a wide-ranging overview of the key elements involved in formulating retirement strategies, drawing on the “Primer on Retirement Income Strategy Design and Evaluation” that was recently released by the SOA. It will cover three types of income objective – income floor, income target and income optimization – and how they imply different retirement strategies along with a different set of metrics to evaluate outcomes. It will also touch on: key member attributes to account for; the nature of income risk; approaches to investing and drawing down on available assets to generate income; and how stochastic modelling can support the analysis. This webcast should deliver something of interest for all attendees, regardless of prior familiarity with strategies for retirement.

Description: Cluster modeling is emerging as a powerful technique to manage cloud costs and run time. It also enables quick responses to "what if" scenarios, and aids in attributing results between reporting cycles. In this webcast, we provide insight into cluster modeling, present case studies that focus on various key performance indicators and discuss the benefits and challenges to cluster modeling from an insurer’s perspective. We also discuss the considerations associated with moving from an on-prem based computing solution to a cloud-based approach.

Description: Mortality Improvement is fundamental to setting your mortality assumptions. Creating your own Mortality Improvement assumption may seem like a Herculean task. The truth is creating a mortality improvement assumption with tools provided by the SOA, or the Continuous Mortality Investigation (CMI) make the process easy. Leaving you with the Herculean task of deciding if the assumption you created fits your product. Expect to learn about the CMI tool and the SOA's Mortality Improvement Model (MIM) tool, how you might get data to use in those tools, and what assumptions you need to make outside of data like what will the MI be in 20 or 30 years? Whether it is the general population, a segment of the general population, or certain causes of death, you will get very different answers.

Description: The frequent use and complexity of today's actuarial projection models have increased the importance of efficient models. Model optimization can reduce both cloud costs and overall end-to-end clock times. This, in turn, enables companies to do more with less, such as reducing close times, increasing controls, enhancing analysis, doing more complicated calculations, and/or improving employee workloads during the close. From design and data to execution and audit, join us to explore techniques that can be applied throughout the actuarial modeling process to optimize actuarial models for improved run-time, accuracy, and reliability. We will focus on techniques that don't rely on model compression.