Search Presentations
STAY TUNED! If you are anticipating additional search filters by attribute and level to align with the CAS Capability Model, it is coming later this Summer. As the CAS begins to code recorded sessions by specific attributes and levels (starting with the 2023 Annual Meeting), these will be tagged in the CAS database of presentations going forward and should be searchable.
But you may use the Capability Model now to help you identify topics. For example, if you want to move up one level under the content area “Functional Expertise,” you may search topics in the particular functional area to expand your knowledge.
Recorded content is searchable by Capability Model attribute and level in the CAS Online Library.
Refine Your Search
The presentation materials are offered in connection with CAS professional education offerings. © 2022 Casualty Actuarial Society. All Rights Reserved. The presentation materials may contain copyrighted content the use of which has not been specifically authorized by the copyright owner. You are permitted to view and print the materials for personal/professional noncommercial research purposes. Except for the foregoing, you agree not to reproduce, distribute, modify, create derivative works, or commercially exploit the presentation materials without prior written permission from CAS. Please direct any copyright permission inquiries regarding use of the presentation materials to acs@casact.org.
STAY TUNED! If you are anticipating additional search filters by attribute and level to align with the CAS Capability Model, it is coming later this Summer. As the CAS begins to code recorded sessions by specific attributes and levels (starting with the 2023 Annual Meeting), these will be tagged in the CAS database of presentations going forward and should be searchable.
But you may use the Capability Model now to help you identify topics. For example, if you want to move up one level under the content area “Functional Expertise,” you may search topics in the particular functional area to expand your knowledge.
Recorded content is searchable by Capability Model attribute and level in the CAS Online Library.
CS9: Bridging the Gap - Pricing and Underwriting
The challenges for an underwriter are no longer restricted to understanding the risks of a single piece of property but to holistically analyze the business portfolio to understand trends and patterns both obvious and hidden.
Underwriters are required to filter through a deluge of data to analyze the performance of the portfolio and the different risks and challenges that a dynamic market place presents. They also need to tap into the local market insights and thereby proactively manage their offerings.
To help underwriters manage these ever-growing expectations, there is a need to leverage advanced big data analytics enabled by geospatial technology to give underwriters the tools required to perform end-to-end underwriting analytics. This approach will provide underwriting teams with better analytical insights to enhance profitability and loss ratios, and to influence customer behavior. It provides insights into data that is integrated across external and internal data sets at the right level of granularity. It would help in the better cause of lose analysis, controlling claims frequency & severity, locate pockets of improvement in portfolio performance.
Such a platform can enable underwriters to perform some of the more complex tasks on a day-to-day basis such as layering weather and event information on carrier’s portfolio and comparing it with loss development patterns from the past in order to identify cause of loss. This platform can also identify significant trends across various dimensions and factors leading to spike in key performance indicators, can drill down to cause-effect relationships of events and losses, and can gather “local intelligence” without the physical presence of agencies in neighborhoods.
Source:
2018 Underwriting Collaboration Seminar
Type:
Concurrent Session
Moderators:
Erin Olson
Panelists:
Yash Yedavalli, Pankaj Parashar
CS14: Underwriting's New Reality – Automated Analytics and Rules Technology Case Study
With increasing competitive pressure and higher consumer expectations, the leading insurance companies are leveraging advanced technology, adopting complex predictive models used in pricing and simultaneously automating underwriting rule execution to minimize the required underwriter touch and decisioning time. This is especially true in the small commercial insurance landscape where straight through processing (STP) is expected to lower operational costs while maintaining or improving profitability and enhancing the customers’ experience. To achieve this, a well-defined process for integration of analytical solutions and rules architecture in the underwriting process that leverages the latest technology and data platform is critical.
In our session, we will discuss typical challenges with models and underwriting rules integration such as misalignment, dispersion and lack of feedback loop. We will present a holistic approach that will help overcome these challenges. This will include a discussion on a portfolio management framework and a well-defined feedback loop. To illustrate how one can operationalize the approach, we will demonstrate a working tool that extracts data from the policy administrative system, executes an integrated predictive models/underwriting rules platform and delivers recommended underwriting actions.
Source:
2018 Underwriting Collaboration Seminar
Type:
Concurrent Session
Moderators:
Brian Mullen
Panelists:
Benny Yuen, Rebecca Freeman
CS6: Risk-Based Strategies For Managing Wildfire Exposure In A Changing Landscape
Wildfire risk was a hot topic in 2017; as the insurance community manages the $9B+ fallout from last year’s record-setting catastrophes, the discussion continues into 2018. Recent publications from the California Department of Insurance highlight the need for carriers to evolve their current view of risk for this peril. A data-driven approach to wildfire risk will allow insurers to reliably manage accumulations and projected losses, mitigating current risk management challenges that are resulting in rate increases and non-renewals for property owners in high-risk areas such as the wildland-urban interface (WUI). Changes in wildfire insurance regulations are being considered for companies writing business in California and throughout the west.
This presentation will explore new solutions for managing wildfire risk, suggesting a probabilistic approach to complement the single-score ratings that are widely used today. The speakers will discuss probabilistic metrics such as exceedance probability (EP), probable maximum loss (PML) and average annual loss (AAL), as well as tools to inform operational decisions based on accumulated risk and fire spread during events. Additionally, the presentation will introduce a collaborative initiative in California to design the first property-level wildfire mitigation standard as a way to measure the efficiency of defensible space and home hardening. The proposed standard is similar to the Florida Wind Mitigation Certification program.
Source:
2018 Underwriting Collaboration Seminar
Type:
Concurrent Session
Moderators:
Mark Westmoreland
Panelists:
Ellie Graeden, Clark Woodward, Yue-Jun Yin
CS8: Underwriter's Acceptance of Predictive Models
Too many times predictive modelers, whether actuaries or not, start first by rummaging through their data to see what may be of interest. Too many times, conference presenters say you should really start with the needs of the end user in mind. And too many times, we still all underestimate the work required to make predictive modeling results valuable because we don’t want to believe that it actually requires the mobilization of so many people and resources.
This session will catalogue the effort required in an attempt to flesh out a process that is more than just “good enough.” From pre-work with the end users to getting the information seamlessly into the workflow, we will brainstorm the steps and hear from a professional underwriter who’s guided the process himself.
Source:
2018 Underwriting Collaboration Seminar
Type:
Concurrent Session
Moderators:
Trevor Leitch
Panelists:
Christopher Cooksey, Jeff Beatty
CS11: Challenging Times: Learnings from the U.K. — Present to Future
Bringing together observations from the large corporate sector and retail motor market, this interactive session explores current issues in pricing as well as how technology is changing the environment for pricing actuaries and regulators.
The first half of the presentation will provide insights from:
1) Analysis of pricing data from the London market and how this is used to deliver risk-based supervision (noting that a significant amount of commercial U.S. business is written through London).
2) Observations from recent pricing reviews conducted on motor insurers in the U.K., including the use of technical versus demand modeling.
The second half of the presentation will focus on the future with an interactive session exploring how technology raises broader issues such as ethics and changes in the insurance value chain. Also included will be observations from the U.K. on insurtech, autonomous vehicles as well as a discussion on ensuring regulation remains fit-for-purpose.
Source:
2018 Underwriting Collaboration Seminar
Type:
Concurrent Session
Moderators:
Susan Bermender
Panelists:
Stefan Claus, Cameron Heath
CS3: Overview of the Private Flood Market
Recent advancements in analytics, coupled with new legislation, capacity in the property insurance market, and increased consumer demand, present exciting opportunities for private flood insurance. Milliman will discuss these changes and provide an update on the National Flood Insurance Program, the largest writer of flood in the United States. Milliman will also demonstrate how private companies can evaluate the feasibility of offering flood insurance, and present the pros and cons of various flood pricing structures used in the private market.
Source:
2018 Underwriting Collaboration Seminar
Type:
Concurrent Session
Moderators:
Erin Olson
Panelists:
Mitchell Waldner, Nancy Watkins
CS16: Predictive Analytics: Why It’s Mostly Garbage and How to Make It Useful
This session will level set the basics of predictive modeling and potential applications in underwriting and pricing individual files or portfolios. We will explore how best to introduce the concept to departments that are not using predictive modeling and overcome barriers such as fear and distrust. We will explore differences in speed of adoption by underwriters of varying experience levels and how to work towards consistency of use. Finally, we will provide some examples of how a department could create a culture and workflow that would encourage adoption and consistency.
Source:
2018 Underwriting Collaboration Seminar
Type:
Concurrent Session
Moderators:
Emmanuel Bardis
Panelists:
Carly Burnham, Ryan Crawford
GS1: Applications of Human-Centered Design in Underwriting
Traditional economic theory posits that people make decisions by rationally weighing together all available information in ways that are economically optimal, based on well-articulated sets of fixed preferences. It’s easy to see how this assumption doesn’t hold up in real-world situations. Behavioral economics studies the psychology of how actual humans — as opposed to idealized “econs” — make decisions, both on the job and in their everyday lives. This session will focus on both the challenges — and the opportunities — resulting from the fact that both insurance underwriters and policyholders are, well, human. Quirks of human psychology imply that underwriting decisions can be improved through the use of algorithms; but they must be designed with the underwriter in mind, rather than being designed to replace human judgment. Then we’ll discuss ways in which choice architecture — another form of human-centered design — can be used to nudge policyholder behavior in economically beneficial and pro-social ways.
Source:
2018 Underwriting Collaboration Seminar
Type:
General Session
Moderators:
Elaine Lajeunesse
Panelists:
James Guszcza, Kimberly Holmes, Kelly Cusick, Matt O'Malley
Effective Collaboration between Actuaries and Data Scientists
With an increased focus on advanced modeling and machine learning, collaboration between Actuaries and Data Scientist is important for project success. In this session we will share benefits, opportunities and pitfalls learned from State Farm’s collaboration on various rating model projects.
Source:
2018 Ratemaking, Product and Modeling (RPM) Seminar
Type:
Concurrent Session
Moderators:
Greg Fears
Panelists:
Jeffrey Rambole, Jeff Kinsey
How to Innovate and get Solutions to Market
We’ll review what is happening in the InsurTech space and review some innovative insurance successes. We will go over a case study of what a successful innovation in the insurance market and what they are providing that the traditional insurers are missing. Then, we’ll review how to develop an innovative product and digitize an existing product. Last, we’ll review the path to regulatory approval in the US. How is the US regulatory process different and how to seek approval across the country in an efficient and successful manner.
Source:
2018 Ratemaking, Product and Modeling (RPM) Seminar
Type:
Concurrent Session
Panelists:
Edward Ford, Sheri Scott
CS18: Developing a Data Strategy that Enables Run-the-Business Excellence while Facilitating R&D
Insurers today compete along cost, service and feature dimensions. Each differentiation dimension is enabled by data. As a result, each insurer needs to develop a tailored data strategy to support the carrier’s competitive differentiation, now and going forward. This is achieved by the following steps: (1) analyzing the changing nature of data (e.g., movement from retrospective, financial measurement data to real-time operational data); (2) working through the challenges to using these new data types — from R&D through front-line, operational users; (3) overlaying product and marketing goals, capabilities and limitations; and (4) building infrastructure to support the future operating model (including data use and analytics), with explicit trace-ability between business needs and technology components.
Source:
2018 Underwriting Collaboration Seminar
Type:
Concurrent Session
Moderators:
Susan Bermender
Panelists:
Tony Beirne, Alan Swan
CS10: Big Data: Regulatory and Compliance Issues
This session will focus on the compliance issues that arise in connection with the insurance industry’s increasing use of big data. The presenter will discuss how new sources of data are being used by industry, and the unique regulatory challenges posed by various types and uses of consumer data. Common pitfalls will be covered, ranging from pricing and underwriting determinations to unfair trade practices and claims handling issues. Regulators’ proposed uses of big data for regulatory oversight purposes will be discussed, such as proposals for regulators’ to pool their resources to better evaluate complex underwriting models. The prospect of an NAIC body taking a centralized role in this proposal will be focused on, as well as the concerns such a development would have for industry.
Source:
2018 Underwriting Collaboration Seminar
Type:
Concurrent Session
Moderators:
Elaine Lajeunesse
Panelists:
Fred Karlinsky, Matthew Fay
CS20: Predictive Analytics: Inferred Behavior and Cyber Risk Quantification
One of the challenges the cyber insurance industry faces today is how to quantifiably measure an organization’s cyber security risk. A key challenge in measuring risk is the interpretation of signals, and the evolving mix of both technology and human behavioral factors makes this especially challenging in quantifying security risks. This discussion will provide insight into how analytics and data science can predict the forward-looking security posture of an enterprise and its potential for suffering a data breach by inferring organizational behaviors from externally observable conditions and correlating these with recent breach events.
Source:
2018 Underwriting Collaboration Seminar
Type:
Concurrent Session
Moderators:
Helene Baril
Panelists:
Ben Goodman, Kimberly Manibusan
GS2: 2017 Natural Catastrophes - A Year to Forget or a Year to Remember and Learn From?
From a natural catastrophe perspective, the year 2017 was extraordinary. In the United States alone, insured losses due to weather and catastrophic events was over $85 billion, according to PCS , with significant losses from multiple severe thunderstorms, hurricanes and wildfires. Given such a big loss year, several important questions have emerged.
• What did each of these perils and major events teach us about catastrophic risk and how to manage it?
• Will we look back on 2017 as a blip in the multi-year lull in global cat losses, a shift away from the quiescent period since 2011, or are we moving into a new normal?
• What can be learned from the devastation to more effectively manage, price and underwrite new/existing business given the events of 2017?
Source:
2018 Underwriting Collaboration Seminar
Type:
General Session
Moderators:
Mark Westmoreland
Panelists:
Eric Robinson, Andrew Siffert
CS5: What You Are Missing: Applications of Vehicle History in Auto Insurance Rating and Underwriting
To evaluate the risk associated with personal auto coverage, insurers have traditionally relied primarily on information about the driver, such as age, gender and marital status; location, such as territory or ZIP code; 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 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.
Source:
2018 Underwriting Collaboration Seminar
Type:
Concurrent Session
Moderators:
Helene Baril
Panelists:
Donald Hendriks
CS19: Hurricanes, Floods and Mortgages: Links between Natural and Default Hazards
Natural hazard events like hurricanes and floods damage homes, create financial hardship on individuals and communities, and can impair a borrower’s ability or willingness to make timely mortgage loan payments. Financial strain to make necessary home repairs and loan payments could be due to inadequate coverage for hazard events (e.g., not carrying flood insurance) by the homeowner, or delays in compensation or government aid. Business interruption could affect employment and income. The loss of home means the loss of homeowner policyholder, and unrepaired homes present new hazards at policy renewal.
This session will explore mortgage payment performance in hurricane-affected areas and investigate default trends pre- and post-event using traditional credit variables, natural hazard risk measures and event-related property damage estimates.
Source:
2018 Underwriting Collaboration Seminar
Type:
Concurrent Session
Moderators:
Randi Woods
Panelists:
Tanya Havlicek, Howard Kunst, Amy Gromowski
CS21: Understanding the Changing Used Vehicle Market
Panelists will discuss the impact on the expanding used vehicle market and its impact on insurance carriers underwriting bottom line. Previously damaged used vehicles have considerably higher claim frequency and loss costs. This new rating/underwriting variable is in use in every state in the country. Many large and small carriers are using this new variable to save one or two points to their company’s bottom line.
Source:
2018 Underwriting Collaboration Seminar
Type:
Concurrent Session
Moderators:
Trevor Leitch
Panelists:
Christopher Whipple, Daniel Pickens
CS2: Easy Tree-sy: An Introduction to Decision Trees
This session will cover the background of and theory behind decision trees, their benefits and drawbacks, a demonstration of fitting decision trees with free software, and example applications of decision trees for data exploration and underwriting analysis.
Source:
2018 Underwriting Collaboration Seminar
Type:
Concurrent Session
Moderators:
Randi Woods
Panelists:
Linda Brobeck, Elaine George
CS12: Leveraging Telematics to Improve Driver Safety and Reduce Risk
From smart phones and smart homes to wearables and connected cars — smart devices are increasingly integrated into our daily lives. Through IoT technologies, these devices are shaping our very culture. The goal for all of us is to leverage this information responsibly to improve the outcomes for our partners, our clients (B2B), and the individuals (B2C) that we serve. FICO, the creator of the industry standard FICO® Score for credit risk and the original provider of credit-based insurance scores, has also created a FICO Safe Driving Score. In partnership with eDriving, the FICO Safe Driving Score leverages telematics captured about an individual’s driving behavior to identify drivers at higher risk for collision. During this session, we will discuss our observations and learnings from our research, demonstrate the importance credible risk scoring, and review how this information can be used to reduce risk by improving driving skills and knowledge. Most importantly, we will discuss how this information can also be used to support and augment the underlying risk assessment that the underwriters and actuaries need to incorporate telematics into their risk models.
Source:
2018 Underwriting Collaboration Seminar
Type:
Concurrent Session
Moderators:
Helene Baril
Panelists:
Jim Noble, Rachel Bell
LL5: Dynamic Special Account Simulation
The dynamic specialty account acquisition model simulates the gains and losses of accounts in a competitive broker-driven market. The model was built to train underwriters and account managers to compete against other insurers to obtain a profitable book. The model pits four teams (insurers) against each other over a period of three trials, where each insurer must choose how to price their existing accounts to retain them plus determine what price to charge to take an account away from another insurer. Each insurer starts with an existing book of one or two specialty accounts. The overall size and profitability of each insurer is the same at the start of the exercise. The winner is determined after three trials (each trial represents an annual period) and the winner is determined by a weighted combination of growth and profit; the weights are determined by the four teams at the start of the exercise and fixed in the model.
Source:
2018 Underwriting Collaboration Seminar
Type:
Learning Lounge
Moderators:
Randi Woods
Panelists:
L. Nicholas Weltmann
Python for Actuaries
This session will introduce the Python programming language. We will review some of the significant libraries that are relevant to actuaries. The session will walk through a machine learning exercise using the scikit-learn package.
Source:
2018 Ratemaking, Product and Modeling (RPM) Seminar
Type:
Concurrent Session
Moderators:
Joshua Merck
Panelists:
Brian Fannin, John Bogaardt
UBI: Public and Proprietary Data Collection, Safety, and Profits
This session will present new research focusing on the impact of data collection through usage based insurance programs, on firm profits, the impact of competition, and fatal accidents. These issues will be considered through the lenses of private industry, academia, and public policy.
Source:
2018 Ratemaking, Product and Modeling (RPM) Seminar
Type:
Concurrent Session
Moderators:
Joshua Merck
Panelists:
Allen Greenberg, Benjamin Shiller
Auto Insurance 2028
With so many changes to auto insurance and the market, how will auto insurance look in the future? A panel of experts from across the industry will discuss the changing market.
Source:
2018 Ratemaking, Product and Modeling (RPM) Seminar
Type:
General Session
Moderators:
Scott C. Anderson
Panelists:
Matthew Moore, Thomas Karol, Alexander Timm
GLM I
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.
Source:
2018 Ratemaking, Product and Modeling (RPM) Seminar
Type:
Concurrent Session
Moderators:
Julie Lederer
Panelists:
Ernesto Schirmacher
Severe Weather - Lessons Learned from 2017
This session will be a discussion of the 2017 hurricane, severe weather and wildfire events, with a particular focus on catastrophe modeling and personal lines pricing issues related to these events.
Source:
2018 Ratemaking, Product and Modeling (RPM) Seminar
Type:
Concurrent Session
Moderators:
David Moore
Panelists:
Kellen Miller, Jeff Waters
