Description: This is a case study session where the speakers will describe a project where a major health system converted its provider distinct reimbursement (i.e., separate physician IPA and hospital capitation) to an integrated reimbursement involving global capitation. The session will discuss the strategic issues considered by the parties and how they evaluated them in their decision process. This session will provide attendees an example of how to approach this type of decision.

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: Often actuarial practitioners are faced with working with data that is less than ideal. The data may be observed with gaps in it, a model may suggest variables that are observed at different frequencies, and sometimes predictive analytic results are very fragile to the inclusion or omission of just a few observations in the sample. Data, particularly big data, are often "messy" and something must be done about it. What is the actuary to do about these very practical matters? These types of questions are what this presentation is intended to address. Concerns on missing data, outliers, nonlinearity, mixed-frequency, cross-correlations are illustrated using real datasets in the Asia-Pacific countries. Some possible solutions to the messy data problems will be discussed.

Description: Data science and predictive analytics are rapidly growing fields and the tools are evolving at a formidable pace. The actuary who wants to be most effective will most likely have to learn at least the basics of more than one programming language, such as R, Python, or Julia, depending on the problem at hand. The Jupyter Notebook App allows you to create interactive notebook documents that are relatively language agnostic, so you only need to learn a single integrated development environment. More importantly, Jupyter Notebooks allows you to easily combine input (code), output (including visualizations), and documentation in a single, shareable, document. Learn how to make better use of your time with these very useful notebooks. This demonstration-based talk will also cover JupyterLab, the next generation of notebooks.

Description: Often actuarial practitioners are faced with working with data that is less than ideal. The data may be observed with gaps in it, a model may suggest variables that are observed at different frequencies, and sometimes predictive analytic results are very fragile to the inclusion or omission of just a few observations in the sample. Data, particularly big data, are often "messy" and something must be done about it. What is the actuary to do about these very practical matters? These types of questions are what this presentation is intended to address. Concerns on missing data, outliers, nonlinearity, mixed-frequency, cross-correlations are illustrated using real datasets in the Asia-Pacific countries. Some possible solutions to the messy data problems will be discussed.

Description: The article recaps existing ideas for assessing credibility and describes questions remaining to be answered.

Description: Predictive models have disrupted the traditional underwriting process. In this session, we will review a decade in the life underwriting of a disruptor – the confluence of actuaries, underwriters and clinicians. In addition, perspectives of other stakeholders will be explored.

Description: This session will review limited fluctuation credibility, and how to apply that approach to a GLM. Alternative credibility approaches through the lens of ASOP 25 will also be discussed.

Description: Actuarial tools now include both open and closed source software; both need a strong model governance framework to mitigate potential sources of model risk and external threats. This presentation will discuss how to develop, install and monitor a model governance framework to safeguard company modeling assets and intellectual capital.

Description: If you’re looking for a quick and easy way to begin exploring data analytics, machine learning, and AI, Microsoft has created a number of Data Science Virtual Machine images which include many of the most popular ML/AI tools pre-installed. In this presentation, we will start from scratch and demonstrate how to get onto the Azure cloud, fire up and connect to a Data Science VM, upload data, and train a machine learning model. While this presentation will use Microsoft’s Azure cloud, many of the concepts should be directly applicable to AWS or GCP.