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.
CS19: A Closer Look at Wildfire Risk Assessment
In this session, learn about the most important attributes to consider in underwriting wildfire risk and how to most effectively evaluate those attributes. Rob Galbraith, Director of Property Underwriting at USAA, and George Davis, Senior Vice President at AIR Worldwide, will discuss the factors that underwriters need to consider for strengthening underwriting and the important role of models in the overall risk management process.
Source:
2017 Underwriting Collaboration Seminar
Type:
Concurrent Session
Panelists:
George Davis, Rob Galbraith
CS18: Lights! Camera! Professionalism!
Come enjoy the acting of some fine fellow professionals who will “perform” several skits involving professional dilemmas. The skits always lead to some lively and educational audience discussions. You will walk away from this session with a better understanding of the ASOPs, codes of conduct and how to apply them when you face your own professional ethical dilemmas. Also included will be commentary pertinent to the CPCU Code of Professional Conduct. This session may provide attendees with Professionalism Continuing Education credits.
Source:
2017 Underwriting Collaboration Seminar
Type:
Concurrent Session
Panelists:
Michael Solomon, J. Fricks, Kimberley Ward
Introduction to R
Pre-registration is required for this workshop. A Laptop is required for this workshop.
Are you one of the millions of actuaries around the world who wish they could draw maps showing the progress of hurricanes, fit generalized linear models to predict sporting events, create complex Monte Carlo simulations, or fit loss distributions? Or maybe you're the sort of person who wants to fit a stochastic reserving model and have a computer display stacks of pretty pictures for you. Do you want one point of access for all of your Excel, text, database and web-based data? Are you fascinated by programmers whose first names begin with the letter "R"?
If that's the sort of person you are, this is the workshop for you. And if none of those people is you, maybe all you want is to be able to get R to work for your computer. We can take care of that too.
R is the single most important technological advance for actuaries since the invention of the pocket calculator. Come see what all the fuss is about.
Requirements for this workshop:
• A functioning laptop (PC or Mac)
• Installation of R and RStudio in advance of the workshop (instructions and support for this will be provided)
• Patience, enthusiasm and a sense of humor (instructions for this are available upon request)
Source:
2017 Ratemaking and Product Management Seminar
Type:
Workshop
Panelists:
Gary Wang, Scott Sobel, Radost Wenman, Emma Li
Capital Allocation and Risk/Reward Decision-Making in an ERM Framework
Pre-registration is required for this workshop.
This workshop will provide a historical primer on the subject and address past and current research of the practical ways in reflecting the cost of capital for ratemaking and decision-making. This primer will discuss aspects as developed by Merton-Perold, Myers-Read, Mango's asset share model, and the RMK procedures. Also considered will be current research as it relates to how actual capital should perhaps be considered in percentile layers and/or traunches, based on the risk metrics considered. We have come a long way in ratemaking since using premium to surplus ratios of 30 years ago. Let's continue the evolution.
A laptop is recommended for participation in this session as we will have hands-on technical sessions 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. Role playing will be involved. Attendees will have access to data in advance.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Workshop
Panelists:
Robert Wolf, Neil Bodoff
Advanced Predictive Modeling
Pre-registration is required for this workshop. A Laptop is required for this workshop.
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 day-long, hands-on, R-based workshop will discuss a variety of statistical learning methods that either refine or serve as complements to the GLM framework. Generalized Additive Models [GAM] will be discussed as a means to capture potentially complex nonlinear relationships.
Multilevel/hierarchical models, ridge regression, and lasso regression will be discussed as a way of achieving shrinkage ("credibility weighting") effects within linear models.
Various forms of tree-based modeling (CART, random forests, boosted trees) will be used to exemplify the use of non-parametric, machine learning, and ensemble methods. Time permitting, other machine learning methods such as support vector machines [SVM] and artificial neural networks [ANN] will also be discussed. Core themes such as the bias-variance trade-off and cross-validation will be woven throughout the presentation. The workshop will assume a working knowledge of the R statistical computing environment and GLM modeling.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Workshop
Panelists:
Mark Jones, Jean-Francois Greeff, Kudakwashe Chibanda, Prashant De
CAS Data and Technology Working Party Report
Members of the CAS Data and Technology Working Party will present findings from their report, which was recently featured in CAS E-Forum. The report discusses how, during the ongoing revolution of data and analytics capabilities, information technology (IT) departments and vendors have embraced “data & analytics,” “big data,” and “data science” as the new frontier for informed decision-making. In the P&C insurance industry, the CAS actuary is uniquely well-positioned to partner with IT to advance the potential of these disciplines to benefit the industry. In order to be a participant in the conversation, however, the actuary must have knowledge of the language, practices, tools and techniques of the technology supporting this revolution.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Jim Weiss, Dominique Yarnell, Peter Bothwell
Severe Weather
Pre-registration is required for this workshop.
This workshop provides educational opportunities to learn both about the science and methods used to build weather models, but also provides useful information on how weather can be used within the business of insurance.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Workshop
Panelists:
Don Windeler, Howard Kunst, Thomas Jeffery, Nancy Watkins, Eric Robinson, Edward Byrns
Navigating the Regulatory Environment Around Usage-based Insurance
Usage based auto insurance has had a mixed reception among US insurance regulators. While most regulators would support UBI for its potential to promote greater fairness in rating, some would object due to the apparent invasion of driver privacy, among other things. Hear from an actuary and a product manager from Verisk with years of experience filing UBI products.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Jared Smollik, Thomas Hess, Isaac Wash
Removing Bias—the SIMEX Procedure
This paper was recently approved for e-Forum publication by the CAS Ratemaking Committee. SIMEX (Simulation-Extrapolation) is a very general technique that helps to correct for bias in estimates caused by errors in measurements of predictors. The method is well established in statistical practice, but seems to not be as widely known in actuarial circles. Using ordinary least squares regression as an example, the method is illustrated using some simple R code.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Thomas Struppeck, Sandra Callanan
An Alternative Approach to Credibility for Large Account and Excess of Loss Treaty Pricing
This paper was recently approved for e-Forum publication by the CAS Ratemaking Committee. It illustrates a comprehensive approach to utilizing and credibility weighting all available information for large account and excess of loss treaty pricing. Excess losses are utilized to modify the severity distribution used to calculate the increased limits factors via a simple Bayesian credibility technique that does not require any specialized software to run. Such an approach considers all available information in the same way as analyzing burn costs, but does not suffer from the same pitfalls.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Sandra Callanan, Uri Korn
Modeling and Analytics in a UBI Setting
This session will discuss considerations related to data handling, variable selection, predictive modeling, performance measurement, and program implementation in usage-based auto insurance (UBI). The session will include a carrier perspective as well as content from the UBI chapter of the upcoming textbook Predictive Modeling Applications in Actuarial Science (Volume 2).
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Gregory Hayward, Jim Weiss, Udi Makov
The Evolution of Classification Rating
Insurance pricing continues to evolve – becoming more sophisticated and predictive. This session will discuss how Personal Lines Auto class plans have evolved and become more granular over time, as techniques and computing power have improved. This session will also present an approach, using a specific case study, for evaluating a modern rating plan using the CAS Statement of Principles on P&C Ratemaking and the Actuarial Standards of Practice. Additionally, we will explore how Usage Based Insurance mechanisms will transform how individual behavior impacts pricing.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Claudine Modlin, Jeffrey Kucera, Irene Bass, Katie DeGraaf
Discussion of ASOP Exposure Draft: Estimating Future Costs for Prospective Property/Casualty Risk Transfer and Risk Funding
The Actuarial Standards Board recently released for comment an exposure draft Actuarial Standard of Practice (ASOP) entitled “Estimating Future Costs for Prospective Property/Casualty Risk Transfer and Risk Funding”. This exposure draft follows exposure and comment on two previous exposure drafts related to Property/Casualty Ratemaking.
With this exposure draft, the scope as well as the title has been revised to clarify that the draft ASOP is limited to estimating the future costs for prospective property/casualty risk transfer and does not include the preparation of rate filings and issues related to the interactions between ratemaking, pricing decisions and rate regulation.
The panelists will cover the key components of the current exposure draft. The areas covered in the draft ASOP include all of the items currently listed in the Considerations Section of the Statement of Principles Regarding Property-Casualty Insurance Ratemaking adopted by the Casualty Actuarial Society in 1988.
Attendees will also be encouraged to provide their comments and suggestions directly to the ASB in their formal process for consideration.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Patrick Woods, Christopher Carlson
Application Of Text Mining In Claims Analytics: A Case Study
This presentation will discuss applications of text mining in predictive modeling. Text mining can be a valuable tool to add insights to data based on free text entry and unstructured information. When text data is considered, analyses like marketing and consumer sentiment, claims analytics, and even pricing can be enhanced.
We will focus on a case study in claims triaging for workers compensation and personal auto lines. Traditional claims triaging requires experienced adjusters to predict which claims require the most attention to minimize losses. To facilitate this process, senior adjusters can manually review claim notes and descriptions. By using modern text mining techniques on claim notes, however, insurers can better triage and see the big picture of their claims.
Text mining is a set of statistical and machine learning tools in the field of natural language processing. These methods examine a collection of documents and systematically identify important words and patterns. Topic modeling, a newer method in text mining, can also be used to predict which claim types are represented in a given claim. By understanding claim types, this advanced analysis can predict, e.g., whether an apparent “fender-bender” is more like a typical single-payment whiplash claim or more like a multi-payment claim with possible nerve damage.
The presentation will focus on applications with intuitive and visual results, with technical info reviewed at a high-level.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Benjamin Williams, William Frierson
Implementing a Predictive Model: Challenges and Considerations for Success
Most of the current focus in data science and analytics is on techniques around data extraction and analysis. And while this focus allows for greater depth of insight into companies’ customers, internal practices, and strategies, the immediate output of the data scientist’s work – no matter how valuable – does not initially translate to bottom line dollars or other measureable impact. The difference between that initial insight, and the ultimate impact it might have, is the success or failure of its implementation. In this way, implementation is an equally important partner to data and technique in ensuring successful analytics impact.
What have your experiences been with IT, claims, underwriters and actuaries when trying to build and deploy a model? This discussion look past the statistics and will highlight ways to ensure successful implementation across four important factors: data, technology, people, and project management.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Kate Jordan, Christopher Zittnan
Easy Tree-sy: Everyday Applications for Decision Trees
This session will cover the background/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, GLM residual analysis, and underwriting analysis.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Daniel Murphy, Linda Brobeck, Peggy Brinkmann
Genetic Algorithms with Applications to Insurance
A genetic algorithm (GA) is a method for solving both constrained and unconstrained optimization problems based on a natural selection process that mimics biological evolution. We give a general overview of GAs and focus on a few specific algorithms with insurance applications: symbolic regression (modeling continuous responses), text mining (looking for higher order n-grams) and differential evolution (used for rating factor optimization, a problem in which the objective function exhibits multiple local minima).
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Eliade Micu
Auto and Home Residual Markets Today: Size, Function and Challenges
The personal lines (auto/home) residual markets look very different today than how they have looked in the past. In private passenger auto, the market is a tiny fraction of what it was in the 1990s, and per policy fixed costs to service this market have increased significantly. In homeowners, coastal residual markets have demonstrated elasticity to respond to storm cycles (or lack thereof), with related changes in industry assessment potential. This session will provide an overview of the mechanisms that currently operate to service the US personal lines residual markets, and provide insight into how reforms may benefit the industry and the consumer.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
James Rowland, Gina Schwitzgebel
Predictive Analytics in Strategy and Decisions – A Tour Around the Globe
Predictive analytics can shape key aspects of insurers’ business strategies and thus is gaining popularity world-wide. Regional diversity in practice, market conditions and regulation is influencing the way analytics is being implemented around the globe. The availability of data, analytical sophistication and the maturity level of customer-centric views can also influence many business problems. These problems can be addressed by predictive analytics, and the efficiency and precision of actuarial decisions. We can learn from the global differences as well as from the commonalities.
This session will discuss various analytical challenges around pricing, underwriting, retention management, customer life time value, etc. and provide real life examples acquired through numerous projects in over 20 countries around the globe.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Reuven Shnaps, Adi Bar-Lev
Concierge Claim Service: The Future of Claims with IoT and Telematics
As we all know, claims are costly and have a big impact on customer satisfaction, not to mention the insurance company’s bottom line. According to InsuranceQuotes.com, even filing one auto insurance claim can cause annual premiums to spike on average 44 percent nationally. Additionally, private passenger auto collision loss ratios have risen to 78.3 percent in 2015. From a customer satisfaction perspective, the dissatisfaction stems from the length of time it takes to be notified of how much the insurer will pay and when the insured actually receives payment. As it relates to property, the impacted customer satisfaction perspective has declined due to the total settlement and service interactions.
IoT, including Telematics, data is changing this archetype by offering holistic claim services. With data from the automakers, alarm companies, and sensor providers, insurers are able to:
• Offer proactive services to avoid claims
• Expedite the claim process by getting ‘black box’ data and prefilling the claim system in order to leverage their preferred partners for tow dispatch, body shop, and car rental
• Understand what is happening in the home and reduce loss by mitigating the claim faster, through water sensors, smoke detectors and door sensors
• Identify fraud
• Reduce claim cycle time
Be sure to attend to learn how this concierge approach can assist you in improving your personal lines claims efficiency and satisfaction by leveraging IoT and Telematics data.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Adam Waldron, Dawn Mortimer, Daniel Sterling
IoT’s Impact on the Distribution, Rating and Underwriting of Personal Auto (Live Streaming)
Until recently, insurers have largely determined car insurance premiums based on characteristics that are proxies for actual driving behaviors. For example, the number of vehicles on the policy and territory are proxies for how much and where the vehicle is driven, respectively. Using actual data that answers when, where, how and how much a vehicle is driven is a much better way to calculate premiums.
While telematics data should play a dominant role in rating algorithms, the data is typically only available after the initial UBI policy has been underwritten and the insured installs the required OBDII sensor or app. While insurers may adjust their traditional rating plans to eliminate obvious overlaps, insurers are still relying heavily on traditional factors to calculate a provisional premium that is subsequently adjusted pending a telematics-based driving score. Auto makers (OEMs), telecommunication companies, mobile handset providers and location-based app providers, etc. (collectively called “data suppliers”) either have or will have this data for their customers. Many of these data suppliers are actively pursuing options of how to best monetize existing data through insurance applications. As these services become available, we expect to see game-changing revisions to how insurers price auto insurance.
This session will focus on the how these data suppliers could fundamentally change the way insurance policies are sold, underwritten and rated. The presenters will highlight some of the benefits as well as the challenges for auto insurers.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Geoffrey Werner, Steven Bishop
Internet of Things for International P&C Business
In the last few years, we are seeing more and more digital disruption to the global insurance industry. Technology often disrupts mature industries and profoundly changes the business models for those who choose to evolve.
In this session, we will briefly discuss lessons from other industries on how disruptive technologies changes the nature of competition and business models, provide an overview of the three core digital technologies (Internet of Things, blockchain and machine learning) and how these technologies will shake the insurance industry in an unprecedented fashion.
Actual examples will be shown to explain their application in customer data aggregation (entity resolution), product development and product pricing. Furthermore, we will discuss how actuaries can reinvent themselves to play a key role in the innovation process.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Lijuan Zhang, Yuxiang Lei, James Dodge
Tutorial on Open Source Text Mining
In 2010 Louise Francis and Matt Flynn wrote a paper, Text Mining Handbook, describing how to do text mining with open source software. The software used in the paper was R and Perl. As a follow-up to the work described in the paper we will show how to use the current open source tools of R and Python to do text mining. Applications will include analyzing text field from claim data and analyzing survey data. As with the paper, we intend to provide a data set along with code that can be used by attendees to begin working with text data. This is intended to be a practical session. It will introduce some of the basic concepts of text mining and then move on to how to use the tools.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Louise Francis, Matthew Flynn
Hospital Professional Liability—How Do We Price a Moving Target?
Over the past several years, the medical community has consolidated from private practices into larger groups, with many physicians seeking employment at hospitals. As a result, Hospital Professional Liability business has taken on an increasing share of Medical Professional Liability business. Hospitals are also going through many changes, including major changes to healthcare delivery, shifts in the judicial environment, and a rapidly growing number of mid-level providers. The ability for actuaries to provide accurate rates is becoming more challenging. In this session, we will summarize the sources of information that actuaries can reference, and identify shifts that are likely to impact professional liability costs for hospitals.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Charles Mitchell, Joshua Youdovin, Timothy Mosler
Bayesian Analysis of Big Data
While Bayesian methods have attracted considerable interest in actuarial science, they are yet to be embraced in large-scaled predictive modeling applications, due to inefficiencies of Bayesian estimation procedures. This paper develops an efficient Markov chain Monte Carlo algorithm using distributed computing on a cluster of computers. The distributed algorithm dramatically boosts the efficiency of Bayesian estimation and expands the scope of applicability of Bayesian methods. The empirical analysis applies a Bayesian hierarchical Tweedie model to a big data of 13 million insurance claim records. The proposed algorithm achieves as much as 65 times performance gain over existing methods in this application. The analysis demonstrates that Bayesian methods can be of great value to large-scaled insurance predictive modeling.
Source:
2017 Ratemaking and Product Management Seminar
Type:
Concurrent Session
Panelists:
Cheng-Sheng Wu, Yanwei Zhang
