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.
Oil Price and Its Impact on Insurance
Energy from oil is becoming increasingly difficult to obtain at reasonable prices. This issue has many implications for insurance companies.
One issue is that high oil prices are closely linked to recessions. Recession can be expected to affect virtually every line of insurance, from auto insurance to mortgage guarantee. Recession can also be expected to have an impact on future investment earnings.
Another issue is that high oil prices stimulate many responses, from drilling in more risky locations, to uses of alternative energy sources such as electric cars and solar panels on businesses and homes, to consumer behavior that changes loss costs such as miles driven and unsafe home heating. All of these marketplace changes demand responses by insurers.
In this session, we will explain these issues and the implications for insurers more fully.
Source:
2011 Annual Meeting
Type:
concurrent
Moderators:
Karen DeToro
Panelists:
Gail Tverberg
Keywords:
Oil Price
Our Estimates Are Cyclical. Are We Being Fooled?
From our research study of the U.S. property-casualty insurance industry, analyzing 30 years of statutory filings across hundreds of companies, we can see a clear and definite cyclical trend in the way our estimates of ultimate loss change over time and at each evaluation. This trend is so clear that it is potentially predictable - or is it? If it's predictable, then does that mean our models are not picking up all of the trend?
Are we "stupid", or is the trend inherently unpredictable?
How does the cyclical trend impact our best estimates?
How does the cyclical trend impact our models of underwriting and reserve risk? If the market is cyclical, then should we be sampling randomly from a distribution, or should there be more of a direction in our sampling?
Source:
2011 Annual Meeting
Type:
concurrent
Panelists:
Donald Mango, Brian Brown, Jessica Leong
Keywords:
statutory filings
Directors and Officers Liability
This session will begin with an update on Directors and Officers (D&O) litigation and enforcement actions. This will include federal securities class actions, actions regarding mergers and acquisitions, and private company litigations. The next section will discuss the D&O Insurance marketplace including pricing, capacity, coverage and enhancements in policies. Analysis by industry type will also be discussed. The final section will discuss the challenges to actuarial pricing and various methodologies for D&O insurance.
Source:
2011 Annual Meeting
Type:
concurrent
Moderators:
Karen DeToro
Panelists:
John Landis, Jacqueline Urban
Keywords:
Directors and Officers Liability, federal securities
Improving Your Presentation Skills Through Improv
Using fundamental exercises from the theater genre of improvisation, attendees will interact with this session's facilitator to examine business communications. The facilitator will guide attendees with suggestions to improve their communication and negotiation skills. In small groups, attendees will participate in improvisational exercises that emphasize the importance of listening in communication.
Attendees will:
Learn to communicate and negotiate more effectively with colleagues and clients;
Gain a better understanding of both the input and output of communication;
Examine their confidence in communicating and negotiating; and
Improve their communication and negotiation skills.
No experience with improvisation is required
Source:
2011 Annual Meeting
Type:
concurrent
Panelists:
Bob Morand
Luncheon with Featured Speaker Anthony Goldbloom
This year's featured luncheon speaker is Anthony Goldbloom. Mr. Goldbloom will speak on the power of competition and its application to the business of analytics and, by extension, actuarial science. He is the founder and CEO of Kaggle. He assists companies with framing modeling tasks as data prediction competitions, ensuring that competitions reflect real-life projects. Before founding Kaggle, Anthony worked in the macroeconomic modeling areas of the Reserve Bank of Australia and before that the Australian Treasury.
Source:
2011 Annual Meeting
Type:
luncheon
Panelists:
Anthony Goldbloom
Integrating External Data into the Decision-Making/Predictive Modeling Process
Gathering appropriate data is one of the foundational steps in any decision-making process. The first presentation will focus on the integration of external data into a territory predictive modeling project. Beyond the usual geo-demographic variables (e.g., population density), other available external data such as Industry loss costs and competitive information will be included in the analysis to better inform the final selections. The second presentation will provide an overview of types of external data available for attrition and elasticity analysis, such as a Pain Index.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Karen DeToro
Panelists:
Serhat Guven, Ronald Zaleski
Enhancing Generalized Linear Models using Rule Induction
This session will focus on how machine learning, and specifically a form of rule induction, can be used to enhance both the process of generalized linear modeling and the resulting models. The emphasis of this session will be on the complementary aspects of having "two tools in the toolbox" when building models. Attention will be given not only to how the use of rule induction changes the signal that can be found, but also to how it changes the context in which the modeling occurs. Case study examples will be shown.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Karen DeToro
Panelists:
Christopher Cooksey
Electronics of Usage Based Insurance
The process of adding a usage-based insurance (UBI) product requires a knowledge of the telematics industry and technology generally foreign to actuaries and other insurance professionals. This session enlists the assistance of industry experts who work for telematics device providers and have developed solutions for insurers to respond to the competitive pressure of UBI or even improve current UBI products. The session will focus on the problems inherent in utilizing electronic devices to track driving behavior and the practical solutions developed to address the insurance needs.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Karen DeToro
Panelists:
Eric Shishko, Mike Carroll, Bob Mathe
A Case Study: Risk Analysis for Australian Auto
A description of the modelling process and results for the risk analysis of the motor portfolio for a major Australian insurer. Components of the process include:
The use of Treenet for variable filtering to reduce 2,500 potential predictors to around 100;
Sampling and stratification strategies to manage data size without sacrificing signal;
The filtering of correlated variables;
Overall measures of model performance;
The fitting of a main effects GLM with key interactions;
Benchmarking the effect of interactions;
A GLM brute force search for interactions followed by iterative Treenet searches;
The need (or not) for geographic and vehicle category overlays.
Although the need for confidentiality limits our ability to show the fine detail of the process, we illustrate key results using a model of claim frequency for at-fault collision claims.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Venkat Veeramani
Panelists:
Richard Brookes
Price Optimization and Statements of Principles on P&C Ratemaking & Classification
There is a wide range of views on how actuaries performing price optimization analyses are complying with the CAS Statement of Principles on Property and Casualty Insurance Ratemaking and the American Academy of Actuaries' Statement of Principles on Risk Classification (the SOPs), and with the Actuarial Standards Boards Standards of Practice (ASOPs). This general session will address specific components of the SOPs and ASOPs, and the way the speakers view how price optimization analyses conform to our professional standards.
Source:
2011 In Focus Seminar
Type:
general
Moderators:
Venkat Veeramani
Panelists:
Chester Szczepanski, Steven Armstrong
Using Novel Data for Vehicle Rating
Traditional vehicle rating is based on grouping similar vehicles together with some adjustment for loss experience. This session will show how detailed information about vehicle characteristics can be used to develop accurate predictions of insurance loss. We discuss types of data available, and the data preparation involved to use the data. We look beyond simple VIN decoding approaches, and introduce text-mining tools in order to bridge disparate formats of key information. We then discuss the modeling process, and the results of the models on actual industry loss data.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Venkat Veeramani
Panelists:
Mark Richards, Shalini Lakshmi
Expanding Analytics through the Use of Machine Learning
Machine Learning, what it is, how it compares to traditional techniques, and how it can change the use of analytics, will be discussed. Machine Learning techniques can produce surprisingly actionable results across a wide array of insurance functions. Many examples illustrating the wide variety of applications will be shown, with the emphasis on the creativity with which an insurer can utilize business information through analytics.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Phil Kentfield
Panelists:
Christopher Cooksey
Effectively Utilizing Vehicle Telematics Data
Vehicle telematics is steadily increasing in prominence with seemingly limitless potential applications in insurance pricing, underwriting, and claims handling. However, in order to successfully incorporate of this wealth of information, the insurance industry must invest not only in new analytic methods but also in developing and maintaining the associated databases. Due to the scale and novelty of the data this may seem to be a daunting undertaking. This session will focus on techniques for organizing a versatile telematics database as the basis for building products and analyses. We will contrast the structure of a telematics database with existing underwriting and actuarial databases, and address some related topics such as privacy concerns.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Phil Kentfield
Panelists:
Avner Freiberger, Colin Sutherland
Auto Insurance Demand Modeling - Analysis of Internet Purchasing Behavior
Over the last decade, insurance companies have become immersed in applications of predictive modeling. One application of predictive modeling that has received a lot of attention but has not been implemented widely as of yet is an analysis of customer conversion, or the likelihood of writing a customer once they contact you and receive a quote. There are a number of challenges to this type of analysis, including availability of data, breadth of data, and the difficulty in measuring elements that are key to the analysis.
To begin to address these difficulties, comscore, Inc. and Pinnacle Actuarial Resources, Inc. have embarked on groundbreaking research to develop insurance demand models, predicting the likelihood of converting insurance customers that receive quotes on the internet. This session will discuss findings from this research and will discuss other planned areas of analysis.
Topics to be discussed will include:
Examples of factors that drive differences auto insurance demand
Discussion of how price differences impact customer purchasing behavior
Application of text mining to the analysis of conversion
Comparison of characteristics of those that are shopping on the internet to those of the general population
How the customer demand changes as they progress through the purchase process
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Jim Hurley
Panelists:
Nick Kucera, Kevin Levitt
Vehicle History: The New Pillar in Auto Policy Pricing
The introduction of vehicle history data into the auto policy underwriting and rating process presents a significant opportunity to refine rates. It has been approved for use in 44 states with filings in several additional states still pending. Initial feedback indicates that vehicle history is the most dramatic change in the auto insurance industry since the introduction of insurance credit scores. This session will cover current applications for vehicle history data in auto policy pricing based upon recent adoption within the industry.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Jim Hurley
Panelists:
David Lackey, Eric Nesson
Professionalism 'Potpourri'
The presence of professionalism in day to day actuarial work is becoming more and more important as it plays a key role in organizational success and in enhancing the quality of our work. As the volume and types of information we produce and use continues to grow exponentially, oftentimes we find ourselves in uncharted territory dealing with difficult and sometime unfamiliar boundaries of professionalism. This session will address how to deal with such difficult situations and will briefly cover a series of professionalism topics including document retention, Code of Professional Conduct, ASOPs, conflict of interest, peer review, emails, privacy and security, etc. The session will provide you with the opportunities to build knowledge, expertise, and grow as a professional.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Jim Hurley
Panelists:
Aquil Ahmed
Measuring Customer Value: A Different Perspective on Price Optimization
Price optimization, the process by which an optimal premium is determined for each risk insured, has received a lot of attention over the past several years. However, for insurance companies considering this approach, defining exactly what this means practically can raise may questions. What exactly am I optimizing? Profit? Growth? Both? What do I do in cases where the optimal premium is substantially different that what I really can charge? How will an approach like this impact other areas of the company?
In price optimization, the focus, by definition is on price. However, focusing on price unnecessarily narrows the lens to the exclusion of a number of inputs to the process as well as potential applications of the results. Insurers can address these issues by changing the focus of these types of analyses to focus not solely on price, but on customer value.
This session will discuss the following aspects of customer value analysis.
Discuss the components of customer value, including profitability, and the probability of attracting, converting and renewing a risk.
Discuss how competition affects the ability to write and retain risks.
Discuss how to quantify the elasticity of demand using the company's own renewal and conversion information supplemented by public information.
Highlight the differences between decisions made using traditional pricing, price optimization, and long-term customer value approaches.
Discuss applications of customer value analyses to marketing, underwriting, pricing, customer service, and claims.
Discuss the practical considerations of implementing customer value concepts.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Jim Hurley
Panelists:
Roosevelt Mosley
Various Approaches to Variable Selection
Variable selection is a very important task in designing a good model. Variable selection gives the modeler the ability to focus on the most important aspects of the data first and NOT be overwhelmed by the enormity of the data itself. However, there is no single algorithm or method that can guarantee us the most optimal set of variables. Thus, variable selection should be made based on many different approaches as opposed to a few.
This session will talk about various approaches to Variable Selection and ways to test if those approaches are giving you reasonable results. Numerous examples will be presented to make the session more meaningful to actuaries.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Jim Hurley
Panelists:
Mark Richards
Usage Based Insurance: A Step by Step Guide for Small and Medium Sized Companies
As information technology continue to advance, the concept of analyzing driving behavior has shifted from science fiction into reality. Leveraging the resources at their disposal, the leading insurance companies have researched and begun to incorporate driving behavior characteristics into their auto insurance products. To a small or midsized company, such product development can seem daunting. The speakers will outline a general strategy that allows small and midsized companies to tap into the power of driving behavior data. We will define telematics and usage based insurance and discuss the data available, the resources needed and challenges and considerations faced when a company looks into developing and implementing usage based insurance plans.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Jim Hurley
Panelists:
Jeff Stempora
Price Optimization with Imperfect Information
Optimized pricing analyses may be daunting not only because of the considerable data and modelling prerequisites, but the assumptions often necessary to compensate for lack of quality information. In this talk we'll discuss some of the more fundamental problems arising in the absence of perfect information. For instance, outside of a randomized price test, elasticity estimates will be biased due to an inadequate explanatory premium term, or so noisy as to be completely unreliable. What alternatives are there for the premium term? What are the symptoms of bias, and how can it be mitigated?
More generally, given that a full, internal solution may not be practical, what are the most important prerequisites to doing a price optimization analysis? What sorts of partial information can be used effectively? Can industry-level information complement missing elasticity estimates, unaccepted quote data, or other common issues, and what are the necessary technical prerequisites to use it?
Source:
2011 In Focus Seminar
Type:
concurrent
Panelists:
Bruce Davey
Peril Interdependencies and Multi-Variate Trend Analysis for Homeowners
Predictive models are used by insurers for underwriting and ratemaking in personal lines insurance. Focusing on homeowners insurance, this presentation provides a systematic comparison of several predictive generalized linear models. We compare pure premium (Tweedie) and frequency/severity models based on single perils as well as multiple perils. With multiple perils, we also introduce instrumental variable models that account for interdependencies among perils. We calibrate these models using a database of detailed individual policyholder experience. We then evaluate the many alternative models in several ways.
We also review how loss trends can be affected by perils trending at different rates, causing changes in peril mix over time. Using homeowners data we show an example of how multivariate trend analysis can be used to analyze homeowners loss trends by peril. Economic conditions may affect all perils simultaneously, leading to correlation in loss trend for different perils. There are times, however, when some perils exhibit trends that are significantly different from other perils. Multivariate stepwise regression can be used to detect data that have significant changes in peril mix or independent trends by peril. This type of analysis can lead to questions of interest for the actuary in regards to trend factor selection, and for the product manager in terms underwriting, product design, or both.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Jim Hurley
Panelists:
James Calton
Increasing Analytical Capabilities in an Organization
The success or failure of modeling projects extends far beyond mastering the technical details. The panel will explore lessons learned from "real life" modeling and analytics projects across multiple lines of business and organizations.
The discussion will contrast "successful" and "unsuccessful" practices related to articulation of project goals, project planning, staffing, data preparation, IT constraints, consultants / development partners, and corporate culture / change management.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Jim Hurley
Panelists:
Elizabeth Riczko
Getting More Out of Your Existing Data
The use of generalized linear modeling and other advanced techniques in insurance pricing has resulted in more accurate and more detailed pricing. Many companies are now looking to 3rd party data for additional signal on the assumption that the identifiable signal in their existing data has been found. This session will provide case studies of how rule induction can be used to identify signal using only currently existing fields. Examples will include cases where signal is found at the coverage level, over and above existing GLM models. The session will also show how other levels (risk or policy) can be used to extract additional information from existing databases.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Jim Hurley
Panelists:
Christopher Cooksey
Social Network Analysis and its Applications to Insurance
Social network analysis [SNA] is the quantitative study of the relations (social, professional, or otherwise) amongst a group of actors such as people or organizations in order to find important patterns. Historically used in fields such as academic sociology, economics, and anthropology, SNA is today a conceptually rich quantitative field that has proven useful in a variety of business and societal settings, such as marketing, public health, and social media analysis. In particular, SNA also potentially provides a significant new dimension along which to analyze insurance policyholders and insurance company employees.
This session will provide some concepts and basic terminology of SNA. In addition, several real-world applications of SNA will be discussed. Finally, various potential uses of SNA in the insurance field will be explored.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Jim Hurley
Panelists:
James Guszcza
Optimal Pricing: Integrating Costs and Customer Behavior
Insurance company financial strategy generally involves goals for growth and profitability. The ratemaking function contributes to these goals through estimation of cost-based rates. That function contributes to financial soundness in the long-term, but in the short- and mid-term ignores the effect that competitive position and customer behavior have on profitable growth. Historically, when these effects have been considered in pricing, it has often been in a subjective and unscientific manner. Insurers can add sophistication to their pricing processes by: developing a better and more granular understanding of competitive position, comparing financial projections of various rate change scenarios, systematic identification of rates that are expected to improve competitive metrics, demand/elasticity modeling and customized price optimization searches.
Source:
2011 In Focus Seminar
Type:
concurrent
Moderators:
Jim Hurley
Panelists:
Claudine Modlin
