Description: This tool created graphs of Google's mobility Data. This version of the tool contains state-level U.S. data through July 31, 2020

Description: Cover a variety of topics centered around the unsupervised learning technique called 'clustering'. The presenters will take you through a wide array of topics that range from 'What is Clustering?' to 'How can I use this technique to solve problems I didn't even know existed?' In the health care industry, everyone is looking for an edge in reducing total spend-cluster analysis finds the unambiguous in an ambiguous framework of results.

Description: Health care has changed rapidly over the past couple of years. The pandemic has shifted consumer behavior in how and where they receive health care as well as consumer attitudes toward health care. Payers continue to design innovative solutions in responses to their patient’s needs and expectations. These changes in consumer behaviors and the responses by payers are expected to impact both health care cost and utilization. Discuss changes in health care that have emerged due to the pandemic and how they are expected to influence actuarial assumptions. Explore epidemiological models commonly used in the industry, the considerations when using these models, as well as the challenges with regards to gathering and interpreting relevant experience data. By attending this session, you will be able to: Identify shifts in consumer behavior that have emerged from the pandemic and describe how payers are responding to new patient expectations. Discuss how these post-pandemic changes by consumers and payers influence medical spending. Determine which actuarial assumptions are impacted and how and if the impact is expected to be temporary or permanent. Discuss considerations taken by actuaries and challenges faced when modelling actuarial assumptions post pandemic.

Description: The COVID-19 pandemic has had an impact on all aspects of daily life and its impact extends to how disability incidence may play out in the future. Product profitability will be shaped by how morbidity patterns, interest rate levels and other key drivers will play out into the future. In this session, our panel will discuss COVID’s impact on IDI morbidity experience as observed over the past 18 months, offer comparisons and insights in light of other insurance lines of business, and share emerging information on the on-going effects of Long COVID. Highlights will also include how regional differences and implications for insurers as the industry moves further into the COVID vaccine era.

Description: Reunite with members and non-members in the China actuarial community to discuss recent topics and related issues in areas such as macroeconomic sharing, IFRS 17, product strategy, cross-border communication and discussion, and more. Advance your career through learning and networking opportunities, and share your experiences in the finance and insurance industries, as well as other fields related to actuarial science.

Description: Level funded products often target the small group market (2-50) and can result in many manual and tedious processes for underwriting to perform risk assessment. Learn how a carrier used technology to help automate risk assessment without the need for paper medical questionnaires, and exposed a risk engine for broker quoting to enable near real-time of pricing for risk-based products. The approach reduced labor and allowed underwriting to focus solely on fallout cases while also improving pricing consistency. Discuss broader trends in the health market for risk assessment, including new machine learning cost predictions that can be used for automation and better predictions of future costs. Discussion topics will include underwriting guidelines, business rules, risk scores, and machine learning. By attending this session, you will have a better idea of new trends in risk assessment for small groups where no claims experience is available. And, better understand business rules that are common in underwriting for flags for review as well as how risk scores and machine learning can be used during the underwriting process.

Description: Watch as presenters highlight the SOA's work on an updated Medicaid underwriting margin model. This research project includes the model along with a set of instructions and a paper describing the key components of underwriting margin, including cost of capital and risk margin. The model was created as a tool for Medicaid actuaries representing both MCO and state principals to determine appropriate underwriting margin components within Medicaid managed care rates. In addition, explore a brief description of the motivation for an underlying theory behind the model. Walk through different model inputs and run through some realistic scenarios that Medicaid actuaries may encounter in states where they operate.

Description: Discuss various analytics and use cases using payer published reimbursement contract data under federal mandate of Transparency in Coverage. These contract rates are publicly available in machine readable format (MRF) starting July 1, 2022, and refreshed monthly. By attending the session, you will have a better understanding of the state-of-art development and analytics using MRF data, including the technologies used to extract and analyze huge MRF files, validation and limitations of published rates, and their use cases.

Description: "Discuss research conducted at the Cleveland Clinic to understand levels of anxiety and depression across Northeast Ohio. Specifically, learn how advanced analytical techniques were utilized with social media data from Twitter and Google to estimate the need for psychological support by geography and overtime. The panel will go further to demonstrate how natural language processing (NLP) and machine learning could be used to make operational decisions at a leading health care provider."

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?