Description: Critical Illness Insurance is one of the fastest growing insurance products available today. Carriers continue to introduce innovative approaches in second, third, and fourth generation products while refining their pricing, underwriting, and claims adjudication processes. This session will look back at historical trends using Gen Re's longstanding, comprehensive Critical Illness Market Survey, and look forward to potential market changes to occur in the next 18 months.

Description: "Visual storytelling is frequently used to effectively convey complex actuarial concepts to our clients and other audiences. When talking about data visualization, we often focus on the explanatory aspect. But data visualization is also a crucial tool in developing insights from large and complex data sets at every stage of data analytics spectrum: • descriptive (what happened?), • diagnostic (why it happened?), • predictive (what will happen?), • and prescriptive (how do I influence what will happen?). In exploratory and modeling stages of data analytics’ project, we let the data tell the story or explore our intuition and hypothesis through data pictures. We demonstrate the power of visual display in healthcare data analytics through vivid examples from the healthcare case studies. These illustrations of data visualization exploratory power range from assessing the quality of data or examining outliers, to exploring variables’ relationships and trends, to hypothesis testing, to guiding in feature selection and models’ comparison and evaluation, to model validation. Throughout these illustrations, we will discuss principles of graphical effectiveness, excellence and integrity."

Description: While medical and pharmacy claims data is a well-established source for healthcare analytics, significant data can be found elsewhere in electronic medical records (EMRs) and in hospital admit, discharge, and transfer (ADT) messages. Increasing availability of bulk EMR data to analyze coupled with more powerful and specialized tools for analyzing it presents an opportunity for actuaries to incorporate these sources in their work. This session will take a look at some of the data available beyond claims in EMRs and HL7 ADT messages, and will discuss case studies, with consideration around structure, ease of use, cleanliness, and tools available for analyzing these data types including natural language processing.

Description: Interest rates steadily declining ever since the Volcker Fed tackled inflation in the early eighties. Ten years of stimulating monetary policy brought new lows. We should have braced ourselves for a return to neutrality, but that concept remains elusive. We will explore structural issues that have brought rates to these low levels and the potential impacts on the insurance business.

Description: With advancement of computing technology and innovative product design, creating dependent and nested stochastic scenarios to accurately predict insurance business has become more feasible and affordable than it used to be. However, can technology and research be accessible, applicable and beneficial to all the stakeholders? Sharing research findings and its validation tools has been our interest and focus, and even a resource-saving strategy. The presenters will share a case study on a segregated fund with hedging strategy. We investigate the reserve calculation for its GMMB in an extensive but simplified stochastic model so that our method is adoptable, applicable and extendable to real-word segregated fund/separate account model. The segregated fund model theory and implementation will be presented and its cash flows will be simulated and analyzed for reserve calculation. Varying parameter assumptions will be tested and to be modified for real-world segregated fund with hedging risk for guaranteed benefits. At the conclusion of the session, attendees will gain basic understanding of and access to: - The option pricing model and its application to the segregated fund (variable annuity's separate account) guarantee - Stochastic scenario generators for the interest rates and equity returns - R-code to program from the scenarios simulations to cash flows of hedging errors - Reserve calculations and its sensitivity to various parameter values and - Comparison of methods to increasing efficiency level and the accuracy

Description: Predictive models can improve the traditional actuarial assumption-setting process. Data science techniques can provide deeper insights into experience through their ability to identify key drivers, rank importance and account for interactions. The presenters will cover ways in which companies can utilize predictive analytics to set assumptions, including the various methods available and practical approaches for implementation. They will also discuss the advantages and challenges of using predictive models for setting assumptions for life, annuity and disability products.

Description: 1) Vision of the applications and development of data analytics in life insurance industry in Asia; 2) How big is the pie and what are the current obstacles insurers are facing; 3) Utilization of data and analytics target operating models as a competitive advantage, with focus on future readiness, controls, efficiency and effectiveness; 4) Potential roadmap ahead: a) Assessment of the current state of maturity levels of data and analytics initiatives; b) Identification of technology and competency gaps; c) Establishment of data integration and management process within BUs, alignment of advanced analytics tools across departments, with centralized data governance, technology structure and decision making framework; d) Definition of criteria and metrics to measure key outcomes and business value added. 5) Case study on agency behavior management: a practical example on how data analytics is being used in practice: (a) key business objectives: improve agent behaviors and maximize productivity with the lowest possible compensation costs; (b) data collection and current state assessment; (c) detailed predictive modeling process, modeling outcomes, and business implications; (d) implementation and monitoring process.

Description: 1) Vision of the applications and development of data analytics in life insurance industry in Asia; 2) How big is the pie and what are the current obstacles insurers are facing; 3) Utilization of data and analytics target operating models as a competitive advantage, with focus on future readiness, controls, efficiency and effectiveness; 4) Potential roadmap ahead: a) Assessment of the current state of maturity levels of data and analytics initiatives; b) Identification of technology and competency gaps; c) Establishment of data integration and management process within BUs, alignment of advanced analytics tools across departments, with centralized data governance, technology structure and decision making framework; d) Definition of criteria and metrics to measure key outcomes and business value added. 5) Case study on agency behavior management: a practical example on how data analytics is being used in practice: (a) key business objectives: improve agent behaviors and maximize productivity with the lowest possible compensation costs; (b) data collection and current state assessment; (c) detailed predictive modeling process, modeling outcomes, and business implications; (d) implementation and monitoring process.

Description: Simple mathematical equations often outperform very complex ones in predictive analytics (PA) situations. This session will cover "why" and "how" to guard against overfitting and other potential pitfalls that might not be intuitive.

Description: Natural Language Process (NLP) techniques can be used to find insights in large unstructured data sets and Word Embeddings are pervasive on a wide spectrum of NLP applications. These word representations improve downstream tasks such as Sentiment Analysis, Recommender System, and Topic Modeling. Learn how NLP transforms the Insurance Industry; and how word embeddings and pre-trained models improve the effectiveness and performance of NLP applications. NLP is increasing in importance for insurance companies. Additional application examples are available in session 36.