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.
And the Winner Is:  How to Pick a Better Model
You have just finished running some data through a predictive modeling package. Now all you need to do is summarize the results, send them along, and you're done, right? Wrong. At the absolute minimum, the modeler should understand and demonstrate the goodness of fit of the model. In most cases, the modeler should also prove that the constructed model provides lift over the existing rating structure. After all, what good is a new model if it cannot outperform the competition?
In this session we will explore, in significant detail, three often overlooked components of the modeling process: measuring goodness-of-fit, assessing lift, and internally validating a predictive model. Model development is usually a major investment. We should make sure our models are performing well to get the best bang for the buck.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Sara Hemmingson
Panelists:
Hernan Medina, Dan Tevet, Anand Khare
Allocating/Attributing Capital- A Hands on Exercise- Part 1
A laptop is recommended for participation in this session.
Join us for part 1 of a hands-on, two-part technical session where the audience will be charged with an exercise in allocating capital to lines of business considering various methods. Participants will work in groups to make strategic recommendations based on their respective results. This session (part 1) will focus on the actual various methodologies in allocating capital.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Sara Hemmingson
Panelists:
Robert Wolf, David Ruhm
Actuarial Modernization and Business Intelligence
Analytics and modelling are exploding, as the business community embraces new information flows and the value of data driven decision making. Actuaries have long held data-centric worldviews, and are well positioned to lead the next generation of insight discovery at insurers. However, at many insurers’ Actuarial Effectiveness is limited by: • Out of date technology
• Labor intensive manual and/or ad hoc data production
• Opaque data presentation
• Cultural resistance to widespread access to actuarial work products
Business Intelligence (BI) applications are at the nexus of these challenges. This session explores the Actuary’s role in driving modernization at the insurer by examining technological and business process BI issues including data management, information presentation, dashboards, and business user training.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Sara Hemmingson
Panelists:
Tony Beirne
Actuarial Model Development: You’re Doing It Wrong
The process of developing actuarial models usually begins in Excel, but ends with the development of another application developed by IT in order to add functionality that is believed to be too difficult (or even impossible) to implement in Excel. Coordinating between IT, actuarial, and underwriting takes time away from testing, which is key to delivering a viable product. But Excel has built-in functionality that allows actuaries to develop rating models that can meet all the needs of pricing actuaries and/or underwriters, once you know how to use it. This session will go over a new method of developing actuarial models and contrast it with the traditional development route, explaining how actuaries can use Excel to download from and upload to databases, manage files and folders, and protect the underlying calculations of a rating model while presenting the user with an interface that includes all the functionality they could expect to get with an application developed by IT.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Sara Hemmingson
Panelists:
Dom Yarnell
A Practical Approach to Variable Selection — A Comparison of Various Techniques
Abstract: Selecting a useful list of variables for consideration in a predictive model is a critical step in the modeling process and can result in better models. Sifting through and selecting from a long list of candidate variables can be onerous and ineffective, particularly with the increasingly wide variety of external factors now available from third party providers. This paper explores a variety of variable selection techniques, applied to frequency and severity models of homeowner insurance claims, developed on a dataset with over 350 initial candidate variables. The techniques are evaluated using multiple criteria, including the predictive power of a resulting model (measured using out-of-sample data) and ease of use. A method based on Elastic Net performs well. Random selections perform as well as some more sophisticated methods, for sufficiently long shortlists.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Benjamin Williams, Gregory Hansen
A Canadian Cat Trick: Flooding and Downpours and Ice — Oh My!
A panel of Canadian actuaries will discuss three of the worst catastrophes to affect their country in recorded history. First and worst were the Southern Alberta floods of June 2013, with insured losses greater than any US event that year. Only a few weeks later, the downtown area of Canada's largest city received five inches of rain in two hours during the afternoon commute. Finally, right before Christmas, Toronto was nailed again with the nastiest ice storm to hit Eastern Canada in 16 years. The presentations will also touch on a wide variety of topics including nuances of the Canadian market and how the landscape has been altered for home and business owners, lingering economic effects, insurance coverage and rate changes, reinsurance implications, etc.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Pat Teufel
Panelists:
Brant Wipperman, Frederick Guillot, Dennis Chua
A Brief Introduction to General Sessionized Linear Mixed Models and General Sessionized Additive Models
This session will provide a brief introduction to General Sessionized Linear Mixed Models and General Sessionized Additive Models. We will discuss the basic assumptions underlying these models and how they vary from classical linear models. We will explore their applications and relative advantages and disadvantages. Graphical images will also be used to explain the concepts with an emphasis on intuition over mathematics. Finally, models will be fit to real data to see what the models produce.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Patrick Hayden
Panelists:
Uri Korn, Eric Mann
Workers Compensation: The Winds of Change—Oklahoma and Its Option
In the face of some of the highest workers compensation costs in the nation, stakeholders in Oklahoma have put forth multiple efforts to reign in system costs in recent years. The panel will present a summary of the key provisions, opportunities, and uncertainties resulting from the latest Oklahoma reform legislation (Senate Bill 1062), which became effective February 1, 2014. The Employee Injury Benefit Act (EIBA), which was implemented as part of the legislative change and functions as an alternative to traditional workers compensation insurance, will be discussed in terms of the statutory and regulatory requirements, its similarities and differences to the opt-out system in Texas, and a summary of some of the filed EIBA plans.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Patrick Hayden
Panelists:
Mark Crawshaw, Cary Ginter
Workers Compensation Trends in the Delivery of Medical Services: There is More Than the Affordable Care Act
The workers compensation sector has continuing concerns over the potential impact of the ACA on medical and indemnity severity. Many observers are unaware that the ACA is only one element in a changing healthcare environment. This session will discuss some important changes in healthcare delivery that began well before the ACA, but which the ACA may accelerate. It will also cover recent NCCI research that assesses the sensitivity of workers compensation medical costs to three key features of the ACA: the focus on the management of chronic disease, comparative effectiveness/best practices, and constrained access due to increased demand from new insureds.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Kendall Williams
Panelists:
Harry Shuford, Len Herk
Workers Compensation — State of the Market
An overview of the current state of the workers compensation line will be presented, including a review of financial results, recent trends, and a discussion of where the line might be headed.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Scott Cederburg
Panelists:
Jay Rosen, Mary Gaillard
Using Text Data From Motor-Vehicle Crash Descriptions to Identify Drivers Under the Influence of Legal or Illegal Drugs
The National Motor Vehicle Crash Causation Survey (NMVCC) database provides crash descriptions for a large sample of motor vehicle accidents. The crash descriptions are an unformatted text data that can extend to several hundred words and are very similar in form and substance to property-casualty claim adjuster notes. In a proof-of-concept demonstration, the NMVCC database will be used to demonstrate the value-added that can be achieved when structured data are supplemented with information that can be extracted from text data (e.g., claim adjusters notes).
Claim adjuster notes and accident descriptions provide a large, untapped source of claim information that is often not captured in structured data. Although there are widely-accepted tests to check whether a driver is under the influence of alcohol, no comparable tests exist for whether a driver may be under the influence of a legal or illegal drug. This session will demonstrate how claim adjuster notes can be used to identify drug use (both legal and illegal types of drugs) that is often not captured at the time of the accident or on conventional reporting forms.
A driver in a motor vehicle crash may be reluctant to admit that drug use impaired their driving. Furthermore, some drivers may have considered the consumption or the amount that was consumed not to be noteworthy. Further, conventional reporting forms may not have the necessary options to capture the information – which may especially be the case with pharmaceutical medications. Finally, conventional structured data capture schemes often do not capture information on a large number of medical conditions (e.g., co-morbidities).
This presentation will demonstrate the differences between identifying General Session concepts in textual form (such as “under the influence”) and taking specific substances (such as “taking Oxycontin”). This presentation will demonstrate how text data can be mined and integrated with structured data to perform predictive analytics. The value of the text data is that the explanatory power of predictive analytics is improved over the case where predictive analytics are limited to structured data.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Scott Cederburg
Panelists:
Philip Borba
Using Collateral Information in Ratemaking
Insurers frequently rely on commercially available solutions to supplement their own data when building predictive rating models. A natural question that arises is this: how does one blend the new (collateral) information with one’s own ratemaking data to create an effective predictive model? This session will summarize recent published research in this field, and then illustrate applications of these ideas to industrywide personal automobile insurance datasets, both simulated and actual.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Karen Adams
Panelists:
John Baldan, Jed Frees
Usage-Based Insurance from a Regulatory Perspective
The number of states with a competitive usage-based insurance (UBI) environment continues to grow. This session will focus on UBI from a regulatory perspective with representation from New Jersey Department of Insurance. We will also discuss the requirements and considerations for filing a UBI program.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Karen Adams
Panelists:
Katie DeGraaf, Carl Sornson
Usage-Based Insurance - The International View
Usage based insurance is becoming more popular world-wide, and many people believe that it is the future of personal auto insurance. However, many challenges remain regarding how to harness the large quantities of data collected by UBI devices -- particularly with respect to storage, computation, aggregation and modelling, as well as how to bridge this new concept with current GLM modelling practices.
This session will discuss what our European and Asian actuarial colleagues are doing, and will provide a more complete and in-depth view of UBI in concept and practice. Some case studies from Europe and Asia, as well as the United States, will be presented.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Karen Adams
Panelists:
Christopher Cooksey, Marcus Looft, Tammy Chen
Update on Latest Vehicle Changes and Safety Enhancements
The panelists will discuss recent changes in automotive design and safety characteristics and the latest developments in rating plans utilized for personal and commercial automobile insurance.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Kim Hazelbaker, LeRoy Boison
UBI Technology: A Plethora of Options
Technology is necessary for a UBI program. However, there are lots of technology options in the marketplace - hard-install, OBD device, smartphone, hybrid OBD/smartphone and manufacturer-embedded. This session will begin with an overview of technology available in the market and then explore the various ways that technology supports the UBI program with data collection, customer experience, and value-added services.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Paul Struzzieri
Panelists:
Ryan Morrison, Christopher Dell
Tree Proximity: The Good, The Bad and The Not So Ugly
Tree Proximity is a fascinating, newly available data point for homeowners insurance. Depending on where you are located in the country, Tree Proximity can be a net benefit or net negative. • Positive: Trees can act as a buffer against wind and hail damage.
• Negatives: Trees can fall on home and cause damage; and trees can act like kindling, spreading fires and bringing more damage to the home.
For the first time ever we have affordable, objective, hyper-local, and hyper-accurate Tree Proximity available through both high-resolution aerial and satellite imagery and proven image-processing algorithms.
Using millions of policy years of data, this talk will break down one way of calculating accurate Tree Proximity values and go over the results of where trees are better and where trees are worse.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Paul Struzzieri
Panelists:
Joe Emison, Arindam Samanta, Guy Seeley
Tour of Various Non-GLM Modeling Techniques
GLMs provide a good framework for modeling, they are well documented, account for basic correlations and are easy to interpret and implement, but they do come at a cost. GLMs struggle when it comes to nonparametric relationships, highly correlated covariates, outliers, and large numbers of covariates. We will take you on a tour of some popular statistical techniques that are being used increasingly such as: • Naive Bayes
• Gradient Boosting
• Non-Parametric Regression
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Erich Brandt
Panelists:
Eliade Micu, Ryan Morrison, Daniel Hernandez-Stumpfhauser
Theory and Applications of Propensity Scoring
This session will feature presentations related to propensity scoring, a technique that has many applications in predictive modeling for property-casualty insurance. The speakers will cover the key motivations underlying the methodology, and then move to case studies and applications. The goal of this session is to make the main concepts understandable to those are unfamiliar with the topic, and also impress upon listeners the importance of the technique.
Mr. Guelman will speak on "Estimating Causal Effects of Insurance Pricing and Marketing Decisions Using Propensity Scores". He will describe how optimal pricing and marketing strategies (such as increasing profits, reducing policy lapses, and crossselling) require the estimation of potential outcomes that result from alternative actions taken by an insurer. While randomized experimentation is General Sessionly the best approach for evaluating competing alternatives, in many practical settings, experimental data is often unfeasible or very costly to obtain. In this context, practical estimation tools from the causal inference literature have proven to be useful. In particular, the propensity score is a device for constructing a matched sample of cases, as they would have likely occurred in a randomized controlled experiment.
Mr. Kim will speak on "Estimating the Effects of Medical Case Management on Workers’ Compensation Severity: Utilization of Propensity Scoring with Nearest-Neighbor Matching". His presentation will illustrate the use of propensity scoring with nearest-neighbor matching to estimate the impact of medical case management (MCM) on workers’ compensation severity.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Erich Brandt
Panelists:
James Guszcza, Leo Guelman, Michael Kim
The Use of Data and Analytics to Successfully Navigate the Changing Face of Customer Acquisition in the P&C Industry
The customer acquisition model in personal lines continues to evolve at a rapid pace. Prospective customers, hit with a barrage of marketing efforts, are demanding a faster, simpler, and more accurate quote-to-bind process resulting in an all-time low of quote to bind conversions.
Leverage machine learning and other modern analytical techniques in combination with non-traditional information sources to provide an accurate quote at the point of initial contact with the consumer minimizing the consumer interface while increasing the integrity of the information on the application for insurance, resulting in the increase of consumer conversions while reducing overall underwriting time and cost.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
John Petricelli, William Ayscue
The Standards and Principles: "They are a Changin'!"
What happened in the comment period for the CAS's statement of principle on P&C ratemaking and the ASB's draft ASOP on P&C ratemaking? On January 31, 2015 the comment period for the Discussion Draft of proposed revisions to the Statement of Principles Regarding Property and Casualty Insurance Ratemaking and the draft Actuarial Standard of Practice (ASOP) on Property/Casualty Ratemaking from the Actuarial Standards Board (U.S.) will conclude.
Join members of the Task Forces responsible for the initial drafting of these important documents to discuss the history that lead to the creation of the documents, the status and the path to adoption.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Matthew Mosher
Panelists:
Shawna Ackerman, Christopher Carlson
The First and Second Laws of Pricing
In an effort to improve profits, most companies have already implemented cost-cutting or reengineering initiatives, and instituted measures to improve productivity. Yet relatively little investment and innovation has gone into the one factor that can have a significantly larger impact on profitability—pricing. Topics we’ll cover include: • Customers don’t buy what, or how, they buy why
• The first law of pricing: all value is subjective
• The second law of pricing: all prices are contextual
• There is no such thing as a commodity
• How people buy
• Pricing Mr. Spock vs. Homer Simpson—the insights of behavioral economics
By gaining an understanding of these principles, attendees will receive a framework that will enable them to better match price to value, as well as experiment with various pricing strategies in order to drive profitability.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Jeffrey Courchene
Panelists:
Ronald Baker
The Actuary in a Predictive Modeling World
Predictive Modeling is a growing discipline in insurance and other industries. Are actuaries poised to play in this ever-changing, fast-paced arena? You will have the opportunity to participate in a hands-on, interactive session. Come prepared to make your voice heard while also hearing from industry experts in this important field.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Jeffrey Courchene
Panelists:
Steven Armstrong, Kimberly Holmes, Alietia Caughron, Sheri Scott
The Actuary as an Expert Witness
This session will introduce a mock expert witness testimony of an experienced ratemaking Actuary. Commentary will be provided by the moderator while the plaintiff's attorney cross-examines the Actuary. Both positive and negative aspects of the testimony will be highlighted and audience participation will be highly encouraged.
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Jeffrey Courchene
Panelists:
Shawna Ackerman, Rich Fidei, Sheri Scott, Timothy Muth
Technology, Entertainment and Design (TED) Talks: Innovation
Since 1984, the annual TED (Technology, Entertainment, and Design) conference features passionate speakers delivering short presentations under the motto – "Ideas Worth Spreading." This year's Ratemaking and Product Management Seminar features three actuaries who feel passionately about innovation. They will share their ideas with the broader actuarial community in three "TED-style" keynote addresses. The insurance industry has been historically slow to change, but the world around us is moving quickly. Our risks are changing, technology is evolving, and customer needs are different today than they were 10 years ago. What will be the next great innovation in insurance? What role can actuaries play in getting us there?
Source:
2015 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Moderators:
Jeffrey Courchene
Panelists:
Claudine Modlin, Roosevelt Mosley, Kevin Mahoney
