2020 Ratemaking, Product and Modeling Seminar Program Guide

The program guide is a list of unique sessions that will be offered at the Seminar. Review session and speaker information for the event as well as any presentation materials that have been submitted. Please check back periodically for any updates as we lead up to the event.

Filtered by:

Clear All
Monday, March 23
2:00pm W-4: Advanced Predictive Modeling Gary Wang, Pinnacle Actuarial Resources, Inc.; Vahid Meimand, Verisk Analytics; Hoi Leung, AIG show more Description Although the Generalized Linear Model (GLM) is a natural foundation for much actuarial work, it is best viewed as a starting point, not the last word on the subject. This full day, hands-on workshop will discuss a variety of statistical learning methods that either refine or serve as complements to the GLM framework. Before the introductions of more advanced predictive modeling techniques, there are some important considerations for more formal modeling studies. The topic of reproducible research, sampling, and hyper-parameter tuning will be discussed and illustrated. Next, you will learn about regression approaches that will expand your toolkit to deal with advanced problems like dealing with large numbers of predictors and fitting nonlinear relationships. We will cover both the theory and practice of using penalized regression and generalized additive models. The concept of "ensembling’ and basics of decision trees as an alternation to GLM will also be introduced individually and predictive models will be built based on combination of the concepts of ensembling and decision trees. All practical examples will be coded in R, with opportunity to work with the code individually or in groups. Attendance will be limited to 40 participants. Time Monday, Mar 23 2:00PM - 10:00PM Location Grand Ballroom B Speakers Gary Wang Pinnacle Actuarial Resources, Inc. Vahid Meimand Verisk Analytics Hoi Leung AIG W-6: Advanced Ratemaking (Half Day-AM) Paul Anderson, Milliman, Inc.; Taralyn Slusarski, Deloitte Consulting, LLP; Daniel Clayman, National Council on Compensation Insurance, Inc.; Robert Moss, National Council on Compensation Insurance, Inc.; Jennifer Beers, Ironshore Insurance show more Description Sometimes basic actuarial ratemaking techniques are insufficient for the job at hand. Data availability, the underlying nature of the exposure, or the presence of other constraints can necessitate the use of additional or more advanced methodologies. This workshop will cover advanced ratemaking topics such as pricing for new products, competitive analysis, large account pricing, and legislative costing. Time Monday, Mar 23 2:00PM - 5:30PM Location Borgne Speakers Paul Anderson Milliman, Inc. Taralyn Slusarski Deloitte Consulting, LLP Daniel Clayman National Council on Compensation Insurance, Inc. Robert Moss National Council on Compensation Insurance, Inc. Jennifer Beers Ironshore Insurance W-1: Basic Ratemaking (Half Day-AM) Lijuan Zhang, AIG; Nickolas Alvarado, Milliman, Inc.; Kevin Hughes, ISO/Verisk show more Description This 1/2 day workshop will explain the fundamental insurance equation: the foundation of the ratemaking process. This session will include techniques for adjusting historical data to estimate its various components, premiums, losses, expenses, and profit, in the relevant pricing time period. The session will conclude by briefly exploring other considerations company management should make, along with the cost-based rate indication derived from the fundamental insurance equation, to determine what rates to charge in practice. This session will also examine the various methods that actuaries use to allocate overall average rates to various subdivisions of a line of business, including territories, classifications, and tiers. Some of the methods discussed will consist of univariate, multivariate, and generalized linear modeling techniques. Time Monday, Mar 23 2:00PM - 5:30PM Location Maurepas Speakers Lijuan Zhang AIG Nickolas Alvarado Milliman, Inc. Kevin Hughes ISO/Verisk W-9: Data visualization and Communication (Half Day-AM) James Weiss, Crum & Forster; Garrett Bradford, Milliman, Inc. show more Description This 1/2 day workshop (morning) will highlight methods for visualizing data in ways that can be communicated to a larger audience. Attendees will have hands-on opportunities to practice these skills Time Monday, Mar 23 2:00PM - 5:30PM Location Southdown Speakers James Weiss Crum & Forster Garrett Bradford Milliman, Inc. W-5: Introduction to Python Brian Fannin, Casualty Actuarial Society; John Bogaardt, WCF Insurance show more Description As actuaries work more and more with teams of data scientists and software engineers, they are more likely to encounter use of the Python programming language. While R is an environment for performing statistical analysis, Python is a general purpose programming language. This workshop will: * provide an introduction to Python; * show how to perform data collection, data visualization, and data analysis in Python; and * point toward resources for further learning. We will be using the Colaboratory platform for instruction. This is a free research platform provided by Google. In order to use Colaboratory, attendees will need a free Google account and an up-to-date browser like Chrome or Firefox. Click this link for a list of FAQs about Colaboratory. If attendees already have Python and Jupyter installed on their devices, they may use this if they wish. A short exercise will be sent out to attendees before the workshop so that they may test the use of the platform. Attendance will be limited to 40 attendees. Time Monday, Mar 23 2:00PM - 10:00PM Location Grand Chenier Speakers Brian Fannin Casualty Actuarial Society John Bogaardt WCF Insurance W-2: Predictive Modeling with Generalized Linear Models Mark Goldburd, Milliman, Inc.; Manyi Luo, Milliman, Inc. show more Description This popular, hands on, interactive predictive modeling workshop has been offered at the RPM seminar for several years. It will be conducted in the R statistical computing language, which is widely adopted by statisticians and data scientists. The course will provide a refresher of regression theory before proceeding to a variety of practical modeling applications and case studies. The focus will be on generalized linear models (GLMs) — their specification, interpretation, and validation. A variety of distributions will be covered, including classical, Poisson, logistic, gamma, and Tweedie. Other topics will be discussed and illustrated as time permits. All datasets will be distributed to attendees prior to the seminar. Attendance will be limited to 40 participants. Time Monday, Mar 23 2:00PM - 10:00PM Location Grand Couteau Speakers Mark Goldburd Milliman, Inc. Manyi Luo Milliman, Inc.
6:30pm W-7: Economic Topics in Ratemaking (Half Day-PM) Gary Venter; Philip Borba, Milliman, Inc.; Stephen Fiete, Aon show more Description This advanced ratemaking workshop will cover a variety of economic topics related to ratemaking, including profit, net cost of reinsurance, as well as the impact of economic conditions on rates. Time Monday, Mar 23 6:30PM - 10:00PM Location Borgne Speakers Gary Venter Philip Borba Milliman, Inc. Stephen Fiete Aon W-8: Executive Presence and Communication (Half Day-PM) Howard Kunst, CoreLogic; Stephanie Rabin, Viridian NYC show more Description This 1/2 day session (afternoon) will allow attendees to learn methods to both create and present communications that can be used at the executive level. Time Monday, Mar 23 6:30PM - 10:00PM Location Southdown Speakers Howard Kunst CoreLogic Stephanie Rabin Viridian NYC W-3: Product Management (Half Day-PM) Paul Anderson, Milliman, Inc.; Todd Witte, Deloitte; Andrew Montano, NCRB show more Description Participants of this workshop will learn how functional areas within an organization work together to make the best decisions for the organizations. The Product Life cycle will also be reviewed. Time Monday, Mar 23 6:30PM - 10:00PM Location Maurepas Speakers Paul Anderson Milliman, Inc. Todd Witte Deloitte Andrew Montano NCRB
Tuesday, March 24
12:00pm MM-3: Coffee and Conversation with the New CAS CEO Victor Carter-Bey, Casualty Actuarial Society show more Description Start your morning by chatting with Victor Carter-Bey, CEO of CAS! At this informal discussion, get to know Victor and share some of your thoughts on recent and upcoming CAS initiatives, as well as the overall CAS member experience. Attendees are invited to grab a cup of joe (or your choice of beverage) and join the conversation. Time Tuesday, Mar 24 12:00PM - 12:45PM Location Grand Ballroom C Speakers Victor Carter-Bey Casualty Actuarial Society MM-2: How Can Actuaries Impact Social Trends Steven Armstrong, Allstate Insurance Company show more Description Join CAS leaders for a discussion around the role of actuaries in relation to current trends impacting our society – for example, on issues such as climate change, technology advancement, aging populations, and affordability of insurance. The roundtable will focus on how actuaries can remain relevant and responsive to change and how the CAS can support this. The discussion will be practice-oriented (and thus count for CE credit), and will also provide discussion and input opportunities regarding CAS direction. Attendees of all experience levels are encouraged to drop by. Time Tuesday, Mar 24 12:00PM - 12:45PM Location Grand Ballroom A Speakers Steven Armstrong Allstate Insurance Company MM-1: Snap! Crackle! Professionalism! Rebecca Williams, North Carolina Rate Bureau show more Description Discuss real-world professionalism scenarios with your fellow actuaries over breakfast. Time Tuesday, Mar 24 12:00PM - 12:45PM Location Grand Ballroom B Speakers Rebecca Williams North Carolina Rate Bureau
1:00pm GS-1: Perspectives on Flood Rebecca Williams, North Carolina Rate Bureau; David Evans, Milliman, Inc.; Patrick Forbes, La. Office of Community Development; David Atkinson show more Description Flood continues to be a topic of national concern, and the response from the private insurance market continues to grow. With a panel of experts in the Flood space, we will explore the current status of the National Flood Insurance Program, provide an update on the Private Market Flood insurance arena, and explore Community responses and efforts related to flood. We will touch on Mitigation efforts, FEMA's federal reauthorization status, and more. Time Tuesday, Mar 24 1:00PM - 2:45PM Location Napoleon Ballroom B1-C3 Speakers Rebecca Williams North Carolina Rate Bureau David Evans Milliman, Inc. Patrick Forbes La. Office of Community Development David Atkinson
3:15pm R-10: Casualty Large Loss Trend (Livestream) Pamela Reale, The Hartford; David Clark, Munich Re America, Inc.; Timothy McCarthy, ISO/Verisk; Peter Magliaro, Zurich North America show more Description Do large losses trend at an inflation rate the same as small losses? This question turns out to be challenging to answer. Naively calculating trend on historical losses above a given threshold is generally useless and at worst misleading. Our session will examine this challenge, including techniques to evaluate the conditions under which a collection of large losses will or will not provide meaningful information. Time Tuesday, Mar 24 3:15PM - 4:30PM Location Napoleon C2 Speakers Pamela Reale The Hartford David Clark Munich Re America, Inc. Timothy McCarthy ISO/Verisk Peter Magliaro Zurich North America M-16: Data Quality Monograph Louise Francis, Francis Analytics & Actuarial Data Mining Inc. show more Description Data is central to the work of actuaries. A significant percentage of the time spent on any analysis is devoted to assembling and cleaning data. With the advent of big data and predictive modeling, even more data is assembled for projects. Data external to the company are increasingly being acquired and used. These external data sources create new challenges for ensuring the quality of the data used in various analyses. This session presents material from a Data Quality monograph being written for actuaries. While the monograph reviews classic data quality concepts, it also introduces newer concepts relevant to data science and predictive modeling that have been popularized by data and data science experts such as Hadley Wickham. This session will have a hands-on focus. It will illustrate the various techniques with public data sources which actuaries can acquire and use. It will also apply open source software (particularly R) that has become popular among data scientists for managing, exploring and cleaning data. Time Tuesday, Mar 24 3:15PM - 4:30PM Location Maurepas Speakers Louise Francis Francis Analytics & Actuarial Data Mining Inc. H-1: Flood Insurance Andrew Montano, NCRB show more Description Discussing flood insurance with your fellow actuaries over breakfast, for CE. All experience (and inexperience) levels welcome! Time Tuesday, Mar 24 3:15PM - 4:30PM Location Napoleon D3 Speakers Andrew Montano NCRB M-18: How Behavioral Economics Can Enable Pro-Social Innovation in Insurance James Guszcza, Deloitte Consulting, LLP show more Description This session will explore a novel way of thinking about “big data” in insurance. In many domains - insurance in particular - applied behavioral economics when paired with behavioral data and digital tech can enable novel pro-social, customer-centric products and business models. Traditionally, behavioral data has been used largely to segment, target, and price risks. In today’s internet of things-connected world, we can also use data to harness the power of data-rich digital environments to nudge safer, less risky, more honest behaviors. The session will provide a brief primer on behavioral economics; discuss how behavioral insights can be used to prompt better driving, safer workplaces, healthier insureds, faster-healing injured workers, more honest claimants, and so on. The panelists will present leading-edge applications of behavioral economics in insurance. Time Tuesday, Mar 24 3:15PM - 4:30PM Location Napoleon C3 Speakers James Guszcza Deloitte Consulting, LLP PIET-7: Innovation Inspired: From Seattle To The Serengeti In Search Of Lessons In Making Changes Kiki Wang, Willis Towers Watson; Craig Thorburn, The World Bank show more Description Can Kenya's insurance challenges learn from Cleveland? Can Melbourne's experience speak to Madagascar? How context can take us from incremental innovation to stepped change? We will showcase examples of how developing countries came up with innovative ideas to solve their unique challenges. From these examples, we will discuss with the audience on what lessons we can learn for US insurance carriers’ own challenges. We will also talk about some existing challenges that the developing countries have and have a group discussion on how we can address them. Through this session, we want to show the audience what’s happening around the world and encourage them to think about innovation outside of their comfort zone. Time Tuesday, Mar 24 3:15PM - 4:30PM Location Grand Couteau Speakers Kiki Wang Willis Towers Watson Craig Thorburn The World Bank PIET-10: Insurance On-Demand: What do we want? Everything! When do we want it? Now! Stephanie Rabin, Viridian NYC; Juemin Zhang, Trov show more Description This session presents firsthand consumer perspectives regarding on-demand insurance products and firsthand actuarial perspectives from developing and pricing such products. Time Tuesday, Mar 24 3:15PM - 4:30PM Location Grand Ballroom A Speakers Stephanie Rabin Viridian NYC Juemin Zhang Trov PIET-5: Insurtech Application in Rate Review and Product Management Automation Yuxiang Lei, Willis Towers Watson; Raul Cisneros, Chubb show more Description An interactive session to explore how to use Insurtech to modernize labor-intensive and time-consuming insurance processes, from rate review data preparation, initial segmentation analysis, regular portfolio monitoring to rate filing preparation, rate change implementation and product management process. Time Tuesday, Mar 24 3:15PM - 4:30PM Location Napoleon D2 Speakers Yuxiang Lei Willis Towers Watson Raul Cisneros Chubb M-14: Intro to Python John Bogaardt, WCF Insurance; Peter Quackenbush, ACCC Insurance Company show more Description As actuaries work more and more with teams of data scientists and software engineers, they are more likely to encounter use of the Python programming language. This session will provide an introduction to Python. Time Tuesday, Mar 24 3:15PM - 4:30PM Location Napoleon C1 Speakers John Bogaardt WCF Insurance Peter Quackenbush ACCC Insurance Company RP-8: Predictive Modeling - A Regulator's Perspective Gennady Stolyarov, State of Nevada Department of Business & Industry, Division of Insurance show more Description Insights on how predictive models are used in the rate filing from the state regulators' perspective. Time Tuesday, Mar 24 3:15PM - 4:30PM Location Grand Ballroom B Speakers Gennady Stolyarov State of Nevada Department of Business & Industry, Division of Insurance R-6: Workers Compensation—State of the Market Jay Rosen, NCCI Holdings, Inc; Rose Barrett, Applied Underwriters show more Description An overview of the current state of the workers compensation line will be presented, including a review of financial results, recent trends, legislative activity, and a discussion of where the line might be headed. Time Tuesday, Mar 24 3:15PM - 4:30PM Location Borgne Speakers Jay Rosen NCCI Holdings, Inc Rose Barrett Applied Underwriters
5:30pm RP-9: 41 Right/ 53 Left/ 38 Right - Unlocking Your Combination Abby Sternberg, Milliman, Inc.; Laura Maxwell, Pinnacle Actuarial Resources, Inc.; Anthony Bustillo, Safety National show more Description Join in the discussion as our panel delves into the considerations associated with ASOPs 41, 53 and 38. Attendees will determine which considerations will be addressed and will be invited to confirm, contradict and otherwise contribute to the panel discussion of the selected considerations Time Tuesday, Mar 24 5:30PM - 6:45PM Location Napoleon C3 Speakers Abby Sternberg Milliman, Inc. Laura Maxwell Pinnacle Actuarial Resources, Inc. Anthony Bustillo Safety National R-11: Don’t Put All Your AALs in One Basket: Event Level Data and Pricing Catastrophic Concentration of Risk (Livestream) Cody Webb, Milliman, Inc.; Garrett Bradford, Milliman, Inc.; John Rollins, Milliman, Inc. show more Description Insurance risk pooling seeks to spread variability across a large number of independent risks. However, for insurers exposed to natural catastrophes, risks are not independent. Significant dependencies exist between individual policies, and surcharges may be warranted for those more correlated with an insurer’s portfolio. In an environment that could include reinsurance cost pressure, property insurers seeking to improve results at the portfolio level may have increased incentive to accurately recognize and allocate reinsurance costs to individual policies. Pricing a concentration charge is a longstanding theme within actuarial literature, but catastrophe models have enabled considerable improvements to traditional techniques. Catastrophe models are most often used to obtain estimates of average annual losses (AAL); but other model outputs at the event level are not as widely used. Actuaries can utilize these model components to measure correlations between individual risks, and calculate concentration charges that are more granular and accurate than most rating plans reflect. In this session, we will review traditional techniques for pricing the concentration charge, describe the data outputs of a catastrophe model, and provide examples of how the event level simulation output could be used by a primary insurer to accurately reflect its reinsurance cost at the territory level, including: 1. What actuaries pricing primary catastrophe insurance need to know about the broader reinsurance market. 2. Why ideal territorial ratemaking approaches for reinsurance cost may differ from those for loss cost. 3. How to conduct a spatial clustering using a K-means algorithm and a Geographic Information Systems (GIS) approach to set reinsurance territories and assess the goodness of their fit. Time Tuesday, Mar 24 5:30PM - 6:45PM Location Napoleon C2 Speakers Cody Webb Milliman, Inc. Garrett Bradford Milliman, Inc. John Rollins Milliman, Inc. R-7: Ensembles of Trees: Could We Actually File One? Christopher Cooksey, Guidewire Software; Brian Poole, Indiana Farm Bureau Insurance Co. show more Description Creating ensembles of multiple decision trees is a common approach in the wider predictive analytics world. These models can be quick to create and highly accurate, but also tend to be very complex. In insurance, a large portion of modeling effort is put toward more accurate pricing. The question then becomes, can ensemble models be filed and used in insurance pricing? We'll cover strategies for incorporating these models into a pricing structure, including using these as submodels and analyzing residuals. We'll also discuss problems and solutions around filing these types of models. Time Tuesday, Mar 24 5:30PM - 6:45PM Location Borgne Speakers Christopher Cooksey Guidewire Software Brian Poole Indiana Farm Bureau Insurance Co. PIET-17: Everything an Actuary needs to know about the Actuarial Climate Index (ACI) and the Actuarial Climate Risk Index (ACRI) Stephen Kolk, Kolkulations LLC; Steven Jackson, American Academy of Actuaries; Steven Armstrong, Allstate Insurance Company show more Description This session will review the content, findings, and outstanding questions posed by both the Actuarial Climate Index (ACI) and the Actuarial Climate Risk Index (ACRI). Time Tuesday, Mar 24 5:30PM - 6:45PM Location Napoleon D2 Speakers Stephen Kolk Kolkulations LLC Steven Jackson American Academy of Actuaries Steven Armstrong Allstate Insurance Company M-6: GLM I Ernesto Schirmacher, Bentley University show more Description Do terms such as “link function,” “exponential family,” and “deviance” leave you puzzled? If so, this session will clarify those terms and demystify generalized linear models (GLMs). The session will provide a basic introduction to linear models and GLMs. Targeted at those who have modest experience with statistics or modeling, the session will start with a brief review of traditional linear models, particularly regression, which has been taught and widely applied for decades. Session leaders will explain how GLMs naturally arise as some of the restrictive assumptions of linear regression are relaxed. GLMs can model a wide range of phenomena, including frequencies and severities as well as the probability that a claim is fraudulent or abusive, to name just a few. The session will emphasize intuition and insight in addition to mathematical calculations. Illustrations will be presented using actual insurance data. Time Tuesday, Mar 24 5:30PM - 6:45PM Location Napoleon D3 Speakers Ernesto Schirmacher Bentley University PIET-8 (R): Insuretech in Commercial Lines Brett Nunes, Oliver Wyman Actuarial Consulting; David Tobias, Betterview; Shobana Sankaran, Esurance; Michael Spinelli, StrongArm Technologies show more Description This session will introduce various Insuretech companies, their offerings and how they expect to change the landscape with how various applicable lines of business will be impacted going forward. We will discuss the impact on various product management and actuarial functions as carriers possibly adapt to leverage this emerging technology. Time Tuesday, Mar 24 5:30PM - 6:45PM Location Napoleon C1 Speakers Brett Nunes Oliver Wyman Actuarial Consulting David Tobias Betterview Shobana Sankaran Esurance Michael Spinelli StrongArm Technologies PIET-16: Pricing, Rating, and Product Personalization – The Game Has Changed Jon Moran; Ben Clark, Frontdoor show more Description In this session, we will explore the topic of personalization – and how the personalization methods and techniques that today’s digital leaders leverage can be applied to the pricing, rating, and product personalization decisions traditional insurers make today. We will see how carriers can customize these pricing, rating, and product variables suit customer needs while meeting organizational goals. We will look at the benefits of insurance personalization - including the maximizing of conversions, revenues, and profits in real time. We will understand how leading insurers are moving towards this model, the use cases they have implemented, and the results that they have seen. The session will conclude by providing five ways that insurance carriers can get started with this new level of personalization – today. Time Tuesday, Mar 24 5:30PM - 6:45PM Location Grand Ballroom A Speakers Jon Moran Ben Clark Frontdoor PIET-9: The Impact of Shifts in Mobility on Insurers Katie DeGraaf, Arity; Ross Reichardt, Arity show more Description Mobility as we know it is changing. Macro-economic shifts have led to an onslaught of rapidly changing dynamics in preferences and risk. These swings manifest within an industry that expertly assesses risk across hundreds of millions of people; insurance. These changes in mobility are demanding changes in strategy to better understand consumers and their risk needs. The industry must evolve by staying ahead of these economic turns while maintaining credibility in assessing risk and serving customers. This discussion will help visualize the economic shifts insurers are facing and steps in order to stay ahead. Time Tuesday, Mar 24 5:30PM - 6:45PM Location Grand Couteau Speakers Katie DeGraaf Arity Ross Reichardt Arity M-15: The Predictive Modeling Cooking Show Richard Sutherland, Travelers; Nathan Hubbell, Travelers; Megan Camanocha, Travelers; Zach Westermeyer, Travelers; Ellie Greiber, Travelers show more Description The “Predictive Modeling Cooking Show” is a window into the kitchen of master chefs as they work, except the chefs are whipping up something different: a predictive model. Join real modelers as they showcase their data engineering and modeling skills, live and on stage. The two teams will go head to head on a classic real-world prediction problem. You'll learn basic modeling concepts including data "scrubbing," the basic steps used to build a model, and how to measure the effectiveness of a model. This is a great chance to gain a better appreciation of modeling concepts in a new and different way. Time Tuesday, Mar 24 5:30PM - 6:45PM Location Grand Ballroom B Speakers Richard Sutherland Travelers Nathan Hubbell Travelers Megan Camanocha Travelers Zach Westermeyer Travelers Ellie Greiber Travelers M-19: Unlocking the Power to Text Data Louise Francis, Francis Analytics & Actuarial Data Mining Inc.; Roosevelt Mosley, Pinnacle Actuarial Resources, Inc. show more Description The amount of information being generated is our society is growing at an astronomical rate, and most of the new information being generated is unstructured. This unstructured data includes text data. While insurance companies have made significant strides in analyzing structured information, there is substantial value that can be extracted from text data. This session will discuss approaches to analyzing text information, and also provide examples of how text information is used in the insurance industry. These examples will include the use of social media information to understand customer sentiment, and analyzing claim adjuster notes to optimize claim outcomes. Time Tuesday, Mar 24 5:30PM - 6:45PM Location Maurepas Speakers Louise Francis Francis Analytics & Actuarial Data Mining Inc. Roosevelt Mosley Pinnacle Actuarial Resources, Inc.
7:15pm M-1: 0 to Git: Open Source for Actuaries Brian Fannin, Casualty Actuarial Society; Rajesh Sahasrabuddhe, Oliver Wyman Actuarial Consulting; Bryce Chamberlin show more Description Open-source projects and tools present new opportunities for collaboration with clients and peers. They can also enhance internal processes through powerful change control and automatic communication. Learn how to use Git processes and understand terms like fork, clone, and branch. See how open-source collaboration works internally and with the external community via issues, forks, merges, and pull requests, including live demos of popular (and free) tools. Learn the benefits and risks to open-sourcing internal projects, how actuaries are participating, and how the profession can benefit by becoming an influencer in the movement. Leave with a detailed Git guide prepared especially for attendees, and the confidence to join the open source revolution! This session will be a prerequisite for the Learning Lab on Git. Time Tuesday, Mar 24 7:15PM - 8:30PM Location Napoleon C1 Speakers Brian Fannin Casualty Actuarial Society Rajesh Sahasrabuddhe Oliver Wyman Actuarial Consulting Bryce Chamberlin PIET-11: Artificial Intelligence in Insurance: Theory and Practice (Livestream) Jason Rodriguez, Willis Towers Watson; Liam McGrath, Willis Towers Watson show more Description Artificial intelligence capabilities and applications continue to expand within insurance. As actuarial and data science teams start to apply these tools to solve business problems, they are continually confronted with the same set of practical challenges and technological limitations. One of the key challenges is model interpretation. In this session, we will introduce two insurance-related AI applications and demonstrate how (and to what extent) we can we understand the relationships captured by these complex models. Time Tuesday, Mar 24 7:15PM - 8:30PM Location Napoleon C2 Speakers Jason Rodriguez Willis Towers Watson Liam McGrath Willis Towers Watson R-5: Auto Insurance: What Happened and What Happens Next? James Lynch, Insurance Information Institute; Marni Wasserman, ISO/Verisk show more Description This session is an update to the Commercial Auto industry experience presented in several recent CAS meetings. Rates have improved significantly in the line, but results continue to be challenging. This session will examine why, from both actuarial and societal points of view. Time Tuesday, Mar 24 7:15PM - 8:30PM Location Grand Ballroom A Speakers James Lynch Insurance Information Institute Marni Wasserman ISO/Verisk PIET-2: Climate Research: A Flooding and Mortgages Case Study David Evans, Milliman, Inc.; Brandon Katz, KatRisk LLC show more Description A changing climate creates a risk to, and opportunity for, actuaries specializing in measuring risk of natural hazards. The use of traditional insurance ratemaking techniques, which typically consider risk over a time horizon of one year, may limit the effectiveness of actuaries in contributing to the growing field of climate change research. By extending the time horizon under study, and broadening the scope of their analysis, actuaries may be able be provide significant value in assessing the financial implications of this risk. This session will present a case study that quantifies the current countrywide exposure to flooding, how it could be impacted by climate change, how this could affect the financial health of residential householders, whether it could impair their ability to meet their mortgage obligations, and what resulting effects might impact other parties. We hope to persuade attendees that actuaries can make significant contributions to climate change research. Time Tuesday, Mar 24 7:15PM - 8:30PM Location Maurepas Speakers David Evans Milliman, Inc. Brandon Katz KatRisk LLC R-2: Cyber Risk and Insurance: How are the Market and Rating Methods Evolving? Danny Arnett, AIR Worldwide; Emma Ye, CyberCube show more Description This session will feature a discussion around the relatively recent emergence of cyber risk and corresponding cyber insurance market, challenges presented to insurers by this new and dynamic landscape, and the intersection of cyber security efforts and traditional insurance practices in helping to address insurer uncertainty in writing cyber risk. The session will also examine current progress and lessons learned from the development of a comprehensive rating plan that can used by insurers and reinsurers to quantify cyber risk. Time Tuesday, Mar 24 7:15PM - 8:30PM Location Grand Couteau Speakers Danny Arnett AIR Worldwide Emma Ye CyberCube M-5: Data Architecture: Do You Know Where Your Data Originated? Meghan Goldfarb, State Farm Insurance Company; Todd Lehmann, Quincy Mutual Fire Insurance Co. show more Description Improper assumptions about data can result in models that don’t deliver the intended outcome, or models that aren’t implementable. Understanding your data architecture – or how the data you use for analytics is captured, managed within the system, and transformed – is critical to properly using that data for analytics and predictive models. This interactive session will introduce key data architectural concepts and provide strategies you can use to make the most of your data. Time Tuesday, Mar 24 7:15PM - 8:30PM Location Borgne Speakers Meghan Goldfarb State Farm Insurance Company Todd Lehmann Quincy Mutual Fire Insurance Co. RP-1: Filing Machine Learning (ML) Based Pricing Models Ralph Dweck, ISO/Verisk show more Description Illustrating the black box model - how to get a machine learning model filed and approved with regulators. Time Tuesday, Mar 24 7:15PM - 8:30PM Location Napoleon C3 Speakers Ralph Dweck ISO/Verisk M-7: GLM II Ernesto Schirmacher, Bentley University show more Description GLM I provided the case for using GLMs and some basic GLM theory. GLM II will be a practical session outlining basic modeling strategy. The discussion will cover topics such as overall modeling strategy, selecting an appropriate error structure and link function, simplifying the GLM (i.e., excluding variables, grouping levels, and fitting curves), complicating the GLM (i.e., adding interactions), and validating the final model. The session will also include a discussion on diagnostics that help test the selections made. Time Tuesday, Mar 24 7:15PM - 8:30PM Location Napoleon D3 Speakers Ernesto Schirmacher Bentley University RP-6: Lights! Camera! Professionalism! Michael Chen, Pinnacle Actuarial Resources, Inc.; Aaron Wright, USAA; Michael Toothman, West Virginia University show more Description Join the Committee on Professionalism as we "act" out scenarios as a method of thinking through situations where the Actuarial Standards of Practice and Code of Conduct may apply in the real world! Time Tuesday, Mar 24 7:15PM - 8:30PM Location Napoleon D2 Speakers Michael Chen Pinnacle Actuarial Resources, Inc. Aaron Wright USAA Michael Toothman West Virginia University R-9: Pricing Emerging Risks Sheri Scott, Milliman, Inc.; Abby Sternberg, Milliman, Inc.; Chin Chang, Hippo show more Description With autonomous vehicles, ride services, smart homes, and the environment we live in rapidly changing, so are the insurance rate development methods. We will walk through how to develop insurance rates when there isn’t sufficient historical loss data on emerging risks to develop the rates. Time Tuesday, Mar 24 7:15PM - 8:30PM Location Grand Ballroom B Speakers Sheri Scott Milliman, Inc. Abby Sternberg Milliman, Inc. Chin Chang Hippo
9:00pm R-1: A Model Based Approach to Personal Auto Geographic Risk Classification Jonathan Fesenmeyer, Allstate Insurance Company; Matthew Berry, Allstate Insurance Company show more Description Geographic (territory) risk classification analyses have some unique challenges – credibility limitations, strong correlations, increased external scrutiny, and the acronym soup of geospatial terms. In this session we’re going to discuss at an approach that combines geospatial analysis with predictive modeling that addresses some of the unique challenges, along with observations that have proved helpful risk classification analysis in general. Time Tuesday, Mar 24 9:00PM - 10:15PM Location Maurepas Speakers Jonathan Fesenmeyer Allstate Insurance Company Matthew Berry Allstate Insurance Company M-10: Advances in Cyber Risk Modeling Ben Goodman, 4A Security & Compliance show more Description Not long ago, the lack of cyber risk data was a major challenge to modeling cyber risk. Today, numerous sources of data are available, but it is still a challenge for insurers to understand how best to use them, given their various limitations. In addition, there are currently several cyber risk models which promise to help insurers quantify cyber risk (frequency, severity & impact), but with a rapidly changing threat landscape, pricing individual and program risks remains a significant challenge. This session will explore current data sources and cyber risk models in the context of the most recent trends in insured cyber losses. We will also take a look at recent advances in the application of AI and machine learning in the quantification of cyber risk. Time Tuesday, Mar 24 9:00PM - 10:15PM Location Borgne Speakers Ben Goodman 4A Security & Compliance RP-3: Back to the Future of Professionalism in Pricing and Predictive Modeling Richard Sutherland, Travelers; Shawn Balthazar, Willis Towers Watson show more Description Great Scott! Predictive modeling is increasingly included in many actuaries’ job descriptions. Whether building predictive models, utilizing them as inputs to the pricing decision-making and implementation process with business partners, or supporting them in rate filings, actuaries are often asked for their opinions on issues that relate to professional ethics. The interactive format remains the same as in previous RPM seminars but the scenarios have been refreshed. We will look Back to the Future of professionalism in pricing and predictive modeling, as we discuss scenarios that touch on common issues related to the Code of Conduct, Statement of Principles, and Standards of Practice. Audience participation is highly encouraged through facilitated discussion and polling. Bring your own DeLorean. Time Tuesday, Mar 24 9:00PM - 10:15PM Location Grand Ballroom A Speakers Richard Sutherland Travelers Shawn Balthazar Willis Towers Watson LL-2: Companion Lab for 0 to Git: Open Source for Actuaries Rajesh Sahasrabuddhe, Oliver Wyman Actuarial Consulting; Bryce Chamberlin show more Description Open-source projects and tools present new opportunities for collaboration with clients and peers. They enhance internal processes through powerful change control and automatic communication. This Learning Lab will give you a chance to get hands-on experience with Git and GitHub after attending the companion session. Go beyond concepts and spend some time actually using these exciting tools. Setup instructions will be supplied. To get the most out of the session, please make sure you have gone through them in advance. Time Tuesday, Mar 24 9:00PM - 10:15PM Location Napoleon C1 Speakers Rajesh Sahasrabuddhe Oliver Wyman Actuarial Consulting Bryce Chamberlin M-8: GLM III Brent Petzoldt, JLT Re; Dominic Dillingham, Oliver Wyman Actuarial Consulting show more Description This session will consider advanced techniques and refinements to the basic GLM which can add material value to the modeling process. It will specifically consider amendments which address some of the purported failings of GLMs in comparison to emerging methods such as machine learning techniques. The session will include a discussion of: • An alternative to driver assignment in personal auto: driver averaging • Innovative ways of modeling bodily injury claims and other miscellaneous refinements • Simple practical modeling steps that can be used to remove distortions created by combining models across claim types • How to tackle the challenging dimension of geography within a GLM • An innovative approach to detecting subtle and higher dimensional interactions in an efficient way, potentially eliminating the need to consider alternative, harder-to-implement-and-explain model forms such as nonlinear models Time Tuesday, Mar 24 9:00PM - 10:15PM Location Napoleon D3 Speakers Brent Petzoldt JLT Re Dominic Dillingham Oliver Wyman Actuarial Consulting RP-2: Has “Hurricane AOB” Passed? The Florida Claims Litigation Crisis and Assignment of Benefits Reform Kathleen Knudson, Milliman, Inc.; Ludi Ye; John Rollins, Milliman, Inc. show more Description Actuarial estimates of expected loss costs often assume that past experience is predictive of the future. However, exposure can be affected by changes in the legal or regulatory environments, and actuaries should account for these changes in their prospective assessments of risk. One area where ongoing changes to the legal environment may be meaningful to actuaries is in Florida, where increasing loss trends arising from “Assignment of Benefits” (AOB) claims have led to legislative remedies via claims litigation reform. In this session, we will provide an overview of the claims litigation environment in Florida, with a focus AOB claims and the potential actuarial implications of the recent AOB reform law (House Bill 7065 in 2019). Time Tuesday, Mar 24 9:00PM - 10:15PM Location Grand Ballroom B Speakers Kathleen Knudson Milliman, Inc. Ludi Ye John Rollins Milliman, Inc. PIET-8: Insuretech in Commercial Lines (Livestream) Brett Nunes, Oliver Wyman Actuarial Consulting; David Tobias, Betterview; Shobana Sankaran, Esurance; Michael Spinelli, StrongArm Technologies show more Description This session will introduce various Insuretech companies, their offerings and how they expect to change the landscape with how various applicable lines of business will be impacted going forward. We will discuss the impact on various product management and actuarial functions as carriers possibly adapt to leverage this emerging technology. Time Tuesday, Mar 24 9:00PM - 10:15PM Location Napoleon C2 Speakers Brett Nunes Oliver Wyman Actuarial Consulting David Tobias Betterview Shobana Sankaran Esurance Michael Spinelli StrongArm Technologies H-2: Keeping it Real about Artificial Intelligence Jason Rodriguez, Willis Towers Watson show more Description Follow-up to concurrent session on AI - opportunity to interact with presenters and fellow attendees! All experience (and inexperience) levels welcomed; counts for continuing education credit. Time Tuesday, Mar 24 9:00PM - 10:15PM Location Grand Couteau Speakers Jason Rodriguez Willis Towers Watson M-3: Machine Learning (ML) Working Party: Practical Methods and Key Issues Navarun Jain, Lux Actuaries & Consultants; James Ely, Convolved, Inc; Marco De Virgilis, Allianz Global Corporate & Specialty SE show more Description The Machine Learning Working Party was formed by the CAS in 2018 to address emerging issues associated with the use of machine learning (ML) techniques in actuarial practice. As part of this effort, the working party has focused on three major areas: 1. Worked examples of ML models 2. Regulatory requirements for communication of ML models (which are often seen as “black boxes”) 3. Survey of ML methods currently in use in the industry This panel focuses on machine learning applications to estimation of pure premiums in automobile liability. Discussion will be centered around an example ML model with comparisons to “traditional” GLMs and Bayesian approaches. We will describe the workings of this model, regulatory considerations in communication, and provide context for other ML models used in the same contexts. Time Tuesday, Mar 24 9:00PM - 10:15PM Location Napoleon C3 Speakers Navarun Jain Lux Actuaries & Consultants James Ely Convolved, Inc Marco De Virgilis Allianz Global Corporate & Specialty SE R-13: The Fully Automated Ratemaking Process Drew Lawyer, Earnix, Ltd.; Julia Hart show more Description Like any business, insurance companies are filled with well-oiled operational processes. The industry has invested millions of manhours refining the underwriting, claims, and customer service processes. We now have advanced straight through underwriting, automated image capture claims payments, and chat bots fielding many customer inquiries. But what about the process of ratemaking? In this session we will take a look at what is needed for a fully automated rate making process. This futuristic view will take us through data management, statistical modeling, rate filing, rate deployment, and performance monitoring. Using advanced scripting language in conjunction with machine learning and other modern statistical techniques, this vision may become a reality within our careers. Time Tuesday, Mar 24 9:00PM - 10:15PM Location Napoleon D2 Speakers Drew Lawyer Earnix, Ltd. Julia Hart
Wednesday, March 25
1:00pm PIET-4: Actuaries as Product Managers - We're Back! Todd Witte, Deloitte; Adam Tyner, Allstate Insurance Company; Quinn Saner, Homesite Insurance; Wasim Chowdhury, Allstate Insurance Company show more Description This interactive session builds upon last year's session, and brings additional perspectives and dialog to engage both new and return attendees. Actuaries build numerous skills through their basic and continuing education and on-the-job training. This panel will discuss how the skills gained and experience as actuaries has helped them transition into roles in the product world. Audience Q&A is welcome! Time Wednesday, Mar 25 1:00PM - 2:15PM Location Grand Ballroom A Speakers Todd Witte Deloitte Adam Tyner Allstate Insurance Company Quinn Saner Homesite Insurance Wasim Chowdhury Allstate Insurance Company R-3: Bayesian Loss Development Methods for Real People Pamela Reale, The Hartford; David Clark, Munich Re America, Inc.; Uri Korn, AIG show more Description This session will compare two strategies for blending loss development patterns. The strategies include curve-fitting and semi-parametric models, but are intended for general practice rather than for a technical audience. The presenters will use live examples to illustrate the methods, with the opportunity for session attendees to follow along with a hands-on example. Time Wednesday, Mar 25 1:00PM - 2:15PM Location Grand Chenier Speakers Pamela Reale The Hartford David Clark Munich Re America, Inc. Uri Korn AIG M-2: Data Visualization for the P&C Actuary Brian Fannin, Casualty Actuarial Society; John Deacon, National Council on Compensation Insurance, Inc.; Jeffrey White, Gross Consulting show more Description The goal of the session is to demonstrate to the audience (P&C actuaries) (1) the benefits of effective data visualization and (2) how to apply proper visualization principles and techniques in their day-to-day work. The visualization principles and techniques presented in this session are based on the research done by the CAS Data Visualization working party. After the overview of data visualization principles and techniques, we will walk through specific examples and then play “Rate That Visual”! “Rate That Visual” is interactive, where we ask the audience to evaluate different visuals for actuarial work (featuring predictive modeling) based upon the principles presented earlier and their own experiences. Time Wednesday, Mar 25 1:00PM - 2:15PM Location Napoleon C3 Speakers Brian Fannin Casualty Actuarial Society John Deacon National Council on Compensation Insurance, Inc. Jeffrey White Gross Consulting M-9: It's Time to Replace MLE with Bayesian Shrinkage for Model Estimation Gary Venter show more Description Fitting models by Bayesian shrinkage instead of MLE improves model accuracy (i.e., it reduces estimation and prediction variances). This previously required ad hoc approaches, but is now a systematic methodology, with clear procedures and a goodness-of-fit measure. It can be applied to most actuarial models, like regression and GLM models, and even reserve development models. This session, coming from a research project done for the Ratemaking Committee, discusses the advantages of both frequentist and Bayesian shrinkage and shows how to apply them. Several classification ratemaking models are fit as illustrations. The session will be a discussion with audience feedback. The steps to do the modeling will be set out, along with an introduction to R packages to implement it. The handouts will have more detail, including code that the attendees can use to experiment and/or modify for their own applications. Time Wednesday, Mar 25 1:00PM - 2:15PM Location Borgne Speakers Gary Venter M-3 (R): Machine Learning (ML) Working Party: Practical Methods and Key Issues Navarun Jain, Lux Actuaries & Consultants; James Ely, Convolved, Inc; Marco De Virgilis, Allianz Global Corporate & Specialty SE show more Description The Machine Learning Working Party was formed by the CAS in 2018 to address emerging issues associated with the use of machine learning (ML) techniques in actuarial practice. As part of this effort, the working party has focused on three major areas: 1. Worked examples of ML models 2. Regulatory requirements for communication of ML models (which are often seen as “black boxes”) 3. Survey of ML methods currently in use in the industry This panel focuses on machine learning applications to estimation of pure premiums in automobile liability. Discussion will be centered around an example ML model with comparisons to “traditional” GLMs and Bayesian approaches. We will describe the workings of this model, regulatory considerations in communication, and provide context for other ML models used in the same contexts. Time Wednesday, Mar 25 1:00PM - 2:15PM Location Maurepas Speakers Navarun Jain Lux Actuaries & Consultants James Ely Convolved, Inc Marco De Virgilis Allianz Global Corporate & Specialty SE LL-1: Presentations, Conversations and Articulations: Public Speaking For (Those Who are Not) Dummies Amy Brener, Casualty Actuarial Society show more Description A large part of our day-to-day professional employment involves communicating with others. While many of us often feel we “live” on email, we do also speak with people face-to-face or over the telephone. We also present at conferences such as this one. This session, loosely based on the successful Toastmasters program, will cover some communication tips and tricks, including guidelines for presenting technical material to non-technical audiences. In addition, those attendees who wish will be given the opportunity to speak “’off the cuff” for 1-2 minutes on a variety of topics. Time Wednesday, Mar 25 1:00PM - 2:15PM Location Grand Ballroom B Speakers Amy Brener Casualty Actuarial Society RP-5: Professionalism Trivia Rebecca Williams, North Carolina Rate Bureau; Lauren Kidwell, GAINSCO show more Description The Committee on Professionalism continues to provide Kahoot! sessions that delve into the Standards of Practice. Will you win with your stellar knowledge of the Standards of Practice? (Bonus: Session is good for Professionalism Continuing Education!) Time Wednesday, Mar 25 1:00PM - 2:15PM Location Napoleon C1 Speakers Rebecca Williams North Carolina Rate Bureau Lauren Kidwell GAINSCO PIET-13: Text Analytics in Claims: Practical Challenges and Lessons Learned Steven Honcharik, Wawanesa Mutual Insurance Company; Yelena Kropivnitskaya show more Description Traditionally insurers treat data-driven decisions as a primary objective for core insurance applications such as ratemaking and reserving. Traditional techniques rely mostly on aggregated, structured data and often leverage purely statistical techniques. Insurers utilize certain types of data, such as risk factors, losses, premiums, and exposures, rating factors, and rates. We have vast quantities of data, both internal and external, that has not been used previously because of its volume, variety, trustworthiness and/or rate of change. Some examples of nontraditional, unstructured data include free-form text notes, documents, images, video files, and call center conversations. Modern data science techniques can help insurers to maximize the use of nontraditional data sources to develop models that drive better business strategies and operational decisions. This talk focuses on text analytics - ranges of technologies for analyzing and processing semi-structured and unstructured text data. Different areas of text analytics are in different stages of maturity and some of them are accessible by practitioners today, while other areas are only now emerging from academia into practice. This talk demonstrates a set of text analytics methods of varying complexity. And illustrates how they are not a substitute for an adjuster’s experience but can be applied to augment the claims process to improve claims operation. The techniques are presented using real world data and model examples. Time Wednesday, Mar 25 1:00PM - 2:15PM Location Napoleon Ballroom C1-C3 Speakers Steven Honcharik Wawanesa Mutual Insurance Company Yelena Kropivnitskaya PIET-6: Wildfire - Modeling and Third Party Liability Issues Tom Jeffery, CoreLogic; Thomas Larsen, CoreLogic show more Description This session will start with a discussion around the science used to estimate Wildfire risk. Understanding the various causes of wildfires (natural vs human) is an important issue that will be discussed; however a major human causation is fires caused by power companies (power lines), and how this third party liability issue will have an impact on insurance going forward. Time Wednesday, Mar 25 1:00PM - 2:15PM Location Grand Couteau Speakers Tom Jeffery CoreLogic Thomas Larsen CoreLogic
2:30pm PIET-14: Are MGAs the future of underwriting? Lon Chang, Ascot Group; Denise Olson, Zurich North America show more Description This session will provide two actuarial perspectives on what MGAs are, how they work, and what the MGA space looks like in the United States. It will explore some situations where a carrier may consider delegating authority and the types of risk typically involved. Case studies of product innovation in alternative markets, oriented towards addressing the needs of the underserved, will be covered. Operational considerations related to MGAs, including compliance and technology, will also be discussed. The session will be useful to anyone interested in tapping into alternative markets, or simply learning how approaches used in alternative markets can apply in traditional ones. Time Wednesday, Mar 25 2:30PM - 3:45PM Location Napoleon C2 Speakers Lon Chang Ascot Group Denise Olson Zurich North America R-2: Cyber Risk and Insurance: How are the Market and Rating Methods Evolving? Danny Arnett, AIR Worldwide; Emma Ye, CyberCube show more Description This session will feature a discussion around the relatively recent emergence of cyber risk and corresponding cyber insurance market, challenges presented to insurers by this new and dynamic landscape, and the intersection of cyber security efforts and traditional insurance practices in helping to address insurer uncertainty in writing cyber risk. The session will also examine current progress and lessons learned from the development of a comprehensive rating plan that can used by insurers and reinsurers to quantify cyber risk. Time Wednesday, Mar 25 2:30PM - 3:45PM Location Napoleon C3 Speakers Danny Arnett AIR Worldwide Emma Ye CyberCube PIET-1: Dive into Private Flood Howard Kunst, CoreLogic; David Evans, Milliman, Inc.; Brandon Katz, KatRisk LLC show more Description Private Flood continues to develop, thanks to recent advances in Flood Modeling. This session will explore private flood programs by looking into locational based risk scores and probabilistic modeling. This session will highlight how modeling has helped bring new private flood products to market. Time Wednesday, Mar 25 2:30PM - 3:45PM Location Grand Ballroom A Speakers Howard Kunst CoreLogic David Evans Milliman, Inc. Brandon Katz KatRisk LLC M-4: Ensembles and Combining Models Christopher Cooksey, Guidewire Software show more Description Ensembles are powerful analytical methods used commonly in the wider Predictive Analytics community outside of Insurance. We'll describe the rationale and explain the effectiveness of Ensembles in general, as well as the basic approaches of bagging and boosting Ensembles of Trees. Advantages of Ensembles on insurance data will be shown and the issues surrounding complex models discussed. Finally, strategies for combining linear regression and ensemble models will be explored. Time Wednesday, Mar 25 2:30PM - 3:45PM Location Maurepas Speakers Christopher Cooksey Guidewire Software H-3: Loss Development Methods for Real People David Clark, Munich Re America, Inc. show more Description This roundtable discussion will expand on the breakout session “Bayesian Loss Development Methods for Real People.” A challenge in implementing Bayesian methods is that prior beliefs need to be translated into explicit statistical distributions. But that is not the way most “real people” think. Our discussion will focus on some ways that we can gain intuition to create useful prior distributions. The potential for Data Science to help may also be explored. Time Wednesday, Mar 25 2:30PM - 3:45PM Location Grand Chenier Speakers David Clark Munich Re America, Inc. M-11: Overview and Practical Application of Machine Learning - Part 1 Benjamin Williams, Willis Towers Watson; Neil Chapman, Willis Towers Watson show more Description The term “machine learning” covers a range of methods that can be powerful, with very practical benefits, in pricing and other insurance applications. Such methods can aid in further improving GLM results or more broadly bring valuable insights to complex problems. There can also be a number of practical challenges in using these methods effectively. This is the first of two sessions reviewing a range of commonly used methods and illustrating different ways they are being applied in insurance, including some case study results. These sessions will focus on the high-level mechanics of each method and the benefits/challenges of their application — as opposed to the underlying technical details. Part 1 will focus on tree-based methods including random forests and GBMs. Part 2 will cover other popular methods including penalized regression methods and neural networks. Each part can be attended without the other. Time Wednesday, Mar 25 2:30PM - 3:45PM Location Grand Ballroom B Speakers Benjamin Williams Willis Towers Watson Neil Chapman Willis Towers Watson RP-4: PIP, PIP, Hooray! The Changing Michigan Auto Market Eric Krafcheck, Milliman, Inc.; Thomas Prince, Milliman, Inc. show more Description Auto insurance coverage in Michigan is among the most generous in the country due to the mandated unlimited Personal Injury Protection (PIP) coverage that every insured must currently carry. However, it comes at a cost, which has long fueled political tensions in the state. In this session, we will discuss legislation that was recently passed by lawmakers (effective in 2020) in an attempt to combat rising auto insurance costs in the state. We will examine the corresponding impacts that the legislation is expected to have on auto insurance in Michigan, including the impact to the Michigan Catastrophic Claims Association (MCCA), as well as the expected rippling effects that the legislation may have on other parts of the economy. Time Wednesday, Mar 25 2:30PM - 3:45PM Location Grand Couteau Speakers Eric Krafcheck Milliman, Inc. Thomas Prince Milliman, Inc. RP-5 (R): Professionalism Trivia Rebecca Williams, North Carolina Rate Bureau; Lauren Kidwell, GAINSCO show more Description The Committee on Professionalism continues to provide Kahoot! sessions that delve into the Standards of Practice. Will you win with your stellar knowledge of the Standards of Practice? (Bonus: Session is good for Professionalism Continuing Education!) Time Wednesday, Mar 25 2:30PM - 3:45PM Location Napoleon C1 Speakers Rebecca Williams North Carolina Rate Bureau Lauren Kidwell GAINSCO R-12: Ratemaking with Telematics Data Victoria Marciano; Jean-Philippe Boucher, University du Quebec a Montreal show more Description In Pay-As-You-Drive (PAYD) automobile insurance, the premium is fixed based on the distance traveled, while in usage-based insurance (UBI) the driving patterns of the policyholder are also considered. In those schemes, drivers who drive more pay a higher premium compared to those with the same characteristics who drive only occasionally, because the former are more exposed to the risk of accident. In this paper, we analyze the simultaneous effect of the distance traveled and exposure time on the risk of accident. Count data distributions based on duration models, where the waiting time or the “waiting distance” between claims are considered will be presented. Semi-parametric models, using Generalized Additive Models (GAM) or Generalized Additive Model for Location, Scale and Shape (GAMLSS) will also be applied using real insurance data. We show how the expected number of claims stabilizes once a certain number of accumulated distance-driven is reached, and propose some explanations to understand this phenomena. Time Wednesday, Mar 25 2:30PM - 3:45PM Location Borgne Speakers Victoria Marciano Jean-Philippe Boucher University du Quebec a Montreal
4:15pm PIET-12: How To Leverage A Robust Decision Making Framework In Developing Product Pricing In Personal Lines Kiki Wang, Willis Towers Watson; Jeffery Hay show more Description As personal lines carriers become more and more sophisticated in analytics, it is important to have a framework in place to understand what the models are telling you, how your underwriting strategies and pricing decisions are affecting your underlying metrics, and most importantly, how different teams - actuaries, data scientists, product managers, underwriters, etc. should work together to create a compelling product. Time Wednesday, Mar 25 4:15PM - 5:30PM Location Grand Ballroom A Speakers Kiki Wang Willis Towers Watson Jeffery Hay R-4: How to use Vehicle History in Today's Ratemaking Donald Hendriks, Carfax; Michael Chen, Pinnacle Actuarial Resources, Inc. show more Description To evaluate the risk associated with personal auto coverage, insurers have traditionally primarily used information about the driver, such as age, gender, marital status and location; the intended use of the vehicle, such as business or pleasure; and information about the vehicle when new, such as original cost new. Insurers then added detail to driver information using sophisticated credit scoring and demographic models. They added detail to vehicle use information with telematics and usage-based insurance pricing. Now, insurers have begun to add detail about the vehicle itself using vehicle history. This session will discuss how carriers use the specific vehicle history to more accurately price and underwrite risks. We will discuss obvious data elements, such as salvage title brands, verified average annual mileage and severe damage. We will also discuss data that may not be as obvious, but has proven predictive of loss in the past. Finally, we will discuss using vehicle history as an entry point for UBI programs Time Wednesday, Mar 25 4:15PM - 5:30PM Location Napoleon Ballroom C1-C3 Speakers Donald Hendriks Carfax Michael Chen Pinnacle Actuarial Resources, Inc. M-13: Implement It and Forget It? No way! Greg Frankowiak, Pinnacle Actuarial Resources, Inc.; Matt Hayden, Liberty Mutual Insurance show more Description Building upon the concept of the modeling lifecycle, this session will take a deeper look at predictive model implementation considerations, including sharing some best practices and potential minefields to look out for. Also, this session will delve into the practice of model monitoring including what it is, why it is so important, and some ways to more effectively ensure that the model that you’ve put so much effort into building is performing the way that you expect it to. Time Wednesday, Mar 25 4:15PM - 5:30PM Location Napoleon D2 Speakers Greg Frankowiak Pinnacle Actuarial Resources, Inc. Matt Hayden Liberty Mutual Insurance M-17: Loss Cost Modeling Using Automated Machine Learning Neal Silbert, DataRobot Inc; Christopher Monsour, DataRobot Inc; Nicholas Hamwey, DataRobot Inc show more Description Loss cost modeling has long been a necessity for insurance companies to set insurance premiums commensurate with the associated risk of each unique policyholder. Since 2000, GLMs have dominated loss cost models, using manually engineered features. With the advent of machine learning, new methods for modeling loss costs have emerged which we will share during this presentation. We will explore the advantages of this new approach: time-saving automated feature engineering; predictions that are not constrained to be a product of factors, but that nevertheless are interpretable; and the resulting business benefits. We will also touch upon claim-level prediction of ultimates and when that is a key input for a loss cost model. Time Wednesday, Mar 25 4:15PM - 5:30PM Location Maurepas Speakers Neal Silbert DataRobot Inc Christopher Monsour DataRobot Inc Nicholas Hamwey DataRobot Inc M-12: Overview and Practical Applications of Machine Learning Methods in Pricing - Part 2 Benjamin Williams, Willis Towers Watson; Neil Chapman, Willis Towers Watson show more Description The term “machine learning” covers a range of methods that can be powerful, with very practical benefits, in pricing and other insurance applications. Such methods can aid in further improving GLM results or more broadly bring valuable insights to complex problems. There can also be a number of practical challenges in using these methods effectively. This is the first of two sessions reviewing a range of commonly used methods and illustrating different ways they are being applied in insurance, including some case study results. These sessions will focus on the high-level mechanics of each method and the benefits/challenges of their application — as opposed to the underlying technical details. Part 1 will focus on tree-based methods including random forests and GBMs. Part 2 will cover other popular methods including penalized regression methods and neural networks. Each part can be attended without the other. Time Wednesday, Mar 25 4:15PM - 5:30PM Location Grand Ballroom B Speakers Benjamin Williams Willis Towers Watson Neil Chapman Willis Towers Watson RP-8 (R): Predictive Modeling - A Regulator's Perspective Gennady Stolyarov, State of Nevada Department of Business & Industry, Division of Insurance show more Description Insights on how predictive models are used in the rate filing from the state regulators' perspective. Time Wednesday, Mar 25 4:15PM - 5:30PM Location Grand Couteau Speakers Gennady Stolyarov State of Nevada Department of Business & Industry, Division of Insurance RP-7: Pricing for Wildfires, Hurricanes, and other Natural Catastrophes - Actuarial Standards of Practice and Regulatory Considerations Sheri Scott, Milliman, Inc.; Chin Chang, Hippo show more Description Many insurer rate plans are outdated and ill-equipped to price for wildfires and other catastrophes that are increasing in frequency. We will walk through how to develop rates using modern ratemaking techniques that consider both Actuarial Standards of Practice as well as state regulatory requirements. Time Wednesday, Mar 25 4:15PM - 5:30PM Location Napoleon C3 Speakers Sheri Scott Milliman, Inc. Chin Chang Hippo PIET-3: Property Valuation Models - How AI and ML are being Used to Improve the Results Taylor Brown, Corelogic; Todd Rissel, e2Value, Inc.; Bin He, CoreLogic show more Description Property valuation models are used across a number of financial services industries, including Insurance and mortgage. This session will describe the various models that are currently used, and how data science is being used to enhance the results. Machine learning and artificial intelligence use case includes image extraction (quality), neighborhood/clustering, and text mining. Time Wednesday, Mar 25 4:15PM - 5:30PM Location Grand Chenier Speakers Taylor Brown Corelogic Todd Rissel e2Value, Inc. Bin He CoreLogic PIET-15: The Road Ahead: Autonomous Trucking and Its Impact on Insurance Andrew Groth, Milliman, Inc. show more Description Autonomous semi-trucks are successfully navigating cross-country routes, and a handful of companies are already using them to deliver goods. Several businesses are considering autonomous trucking as a solution to supplement growing driver shortages and to create more efficient distribution networks. The insurance industry has started to adapt to autonomous vehicles for personal use, but the industry does not have a cohesive approach to address the impact on commercial lines. This revolutionary technology is gaining momentum and will reshape traditional lines of business including auto, workers’ compensation, general liability, and inland marine. It will also bring new liabilities to light that are not typically considered in today’s insurance policies. Actuaries need to gain an understanding of this evolving technology in order to begin quantifying the impact autonomous trucks will have on company stakeholders and to create effective strategies to price and underwrite the risk in the near future. Time Wednesday, Mar 25 4:15PM - 5:30PM Location Napoleon C1 Speakers Andrew Groth Milliman, Inc. R-8: Workers Compensation Experience Rating—2020 and Beyond Yuchen Su, National Council on Compensation Insurance, Inc.; Dave Bellusci, W.C. Insurance Rating Bureau of California show more Description Experience Rating (ER) for workers compensation insurance started more than a century ago, long before predictive modeling became popular for the industry. Over the years, NCCI periodically reviewed its ER Plan and identified ways to refine its capability of using risks’ historical losses to predict their future losses. NCCI’s most recent research began in late 2017 and is the first ER Plan review since the increase of the split point was introduced in 2013. In this session, NCCI will discuss ideas being explored that may improve the performance of the its ER Plan. You will gain glimpses of interesting findings from the current research and have opportunities to pose questions. California’s experience rating plan has gone through a number of transformations over the last several years with the use of variable split points and the exclusion of the first $250 of every loss. This session will also discuss how the recent changes in California are working for system stakeholders and what changes may occur in the future. Time Wednesday, Mar 25 4:15PM - 5:30PM Location Borgne Speakers Yuchen Su National Council on Compensation Insurance, Inc. Dave Bellusci W.C. Insurance Rating Bureau of California
6:15pm GS-2: Avoiding an Extinction Event: Evolution to Artificial Intelligence in the Property & Casualty Insurance Industry Severence MacLaughlin, Capgemini Invent show more Description With the evolution of Big Data and a digital awakening across the global market place the new tech buzz word is Artificial Intelligence. It is on the tip of everyone’s tongue, but is there substance behind the marketing machines that have brought to market AI personas such as Watson, Amelia, Holmes & Einstein? Different market sectors have started to evolve in the adoption of Artificial Intelligence at different speeds with financial services in terms of banking and investments as being one of the quickest to invest BUT deriving a low ROI. Why is this? Is it better to be the tortoise or the hare in the AI race? What is the right way to invest in AI as an Insurance carrier? Does it change based on the size of the portfolio insured? How should the actuarial department look to enhance its capabilities and role within the company? Does the company have the leadership capabilities to ensure that the investment drives an ROI and is not a science project? These topics and more will be discussed. Time Wednesday, Mar 25 6:15PM - 7:30PM Location Napoleon Ballroom C1-C3 Speakers Severence MacLaughlin Capgemini Invent
M-13: Implement It and Forget It? No way! Greg Frankowiak, Pinnacle Actuarial Resources, Inc.; Matt Hayden, Liberty Mutual Insurance show more Description Building upon the concept of the modeling lifecycle, this session will take a deeper look at predictive model implementation considerations, including sharing some best practices and potential minefields to look out for. Also, this session will delve into the practice of model monitoring including what it is, why it is so important, and some ways to more effectively ensure that the model that you’ve put so much effort into building is performing the way that you expect it to. Speakers Greg Frankowiak Pinnacle Actuarial Resources, Inc. Matt Hayden Liberty Mutual Insurance
CLOSE
Day


Time
Track








Session Type



Level of Knowledge