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.
Actuarial Modernization Trends: How Modern is Your Actuarial Function?
Robotic process automation, AI, machine learning, advanced analytics, data infrastructure, visualization – all the latest buzz in actuarial modernization. How far along is the P&C industry on key actuarial modernization initiatives? In this session, we will highlight these key initiatives, presenting findings from PwC’s 2019 global actuarial modernization survey, and deep dive into two areas of focus that are gaining traction in the actuarial space - data visualization and robotics process automation. We will demonstrate a dynamic visualization and an automated reserve analysis before discussing the practical considerations in implementing these tools within your own processes. By the end of the session, you will be able to evaluate where your company stands compared to the industry on various actuarial modernization initiatives. Furthermore, you will be able to assess which of your actuarial processes may be ripe for dynamic visualization or automation and how to get started on the pr
Source:
2020 Spring Meeting
Type:
Concurrent Session
Moderators:
Joe Milicia
Panelists:
Zachary Martin, Elizabeth Casazza McGrew, Sumaali Chheda, Kimberly LaRiccia
Reserving with Machine Learning: Innovations from Loyalty Programs to Insurance
How is machine learning used for loyalty programs and what can it teach us about insurance claim reserving?
While triangular methods have been a foundational tool for decades, individual claim reserving gives the actuary far more information about changes and trends in the liability. Yet the commonly used individual claim reserving techniques leave some of the most valuable data unexamined.
In this session, we’ll cover the benefits of reserving at the individual claim level and describe an approach that sits at the intersection of data science and actuarial science. This session will also introduce a new actuarial tool – the snapshot date triangle – and demonstrate how it can be combined with machine learning to produce a robust and powerful individual claim reserving system. You will learn why the snapshot date triangle was originally developed for estimating loyalty program liabilities and how it can be used in insurance contexts.
Source:
2020 Spring Meeting
Type:
Concurrent Session
Moderators:
Joe Milicia
Panelists:
Len Llaguno, Julie Hagerstrand, Dylan Reed
Estimation of Individual Claim Liabilities: Traditional and Machine Learning Methodologies
The intent of the session is to show and to inform the audience on how to implement machine learning (ML) algorithms in the framework of calculating Claim Liabilities.
Traditionally, in order to estimate future losses, actuaries have been using methodologies based on aggregated data in the form of run-off triangles.
The aim of the presentation is to outline the limitations of such methodologies and propose more sophisticated tools and models based on ML algorithms that are capable of overcoming drawbacks of standard approaches, namely, accuracy and timeliness of estimates.
We propose a new framework that could enhance traditional estimates in providing an additional evaluation tool that could be used by actuaries as another term of comparison.
In addition we will present results achieved alongside advantages and disadvantages, ie. issues that may be encountered in a corporate production environment when transitioning from standard approaches to more sophisticated ones.
Source:
2020 Spring Meeting
Type:
Concurrent Session
Moderators:
Keith Palmer
Panelists:
Linda Brobeck, Marco De Virgilis
The Road Ahead: Autonomous Trucking and Its Impact on Insurance
Autonomous semi-trucks are successfully navigating cross-country routes, and a handful of companies are already using them to deliver goods. Several businesses are considering autonomous trucking as a solution to supplement growing driver shortages and to create more efficient distribution networks. The insurance industry has started to adapt to autonomous vehicles for personal use, but the industry does not have a cohesive approach to address the impact on commercial lines. This revolutionary technology is gaining momentum and will reshape traditional lines of business including auto, workers’ compensation, general liability, and inland marine. It will also bring new liabilities to light that are not typically considered in today’s insurance policies. Actuaries need to gain an understanding of this evolving technology in order to begin quantifying the impact autonomous trucks will have on company stakeholders and to create effective strategies to price and underwrite the risk in the n
Source:
2020 Spring Meeting
Type:
Concurrent Session
Moderators:
Keith Palmer
Panelists:
Drew Groth
Data Science and Improving Claims Customer Experience
A vast array of new technologies along with more advanced data fusion methods now allows refined claims segmentation that optimizes customer journey maps across an expanding set of claims experiences.
Source:
2020 Spring Meeting
Type:
Concurrent Session
Moderators:
Kudakwashe Chibanda
Panelists:
Marty Ellingsworth , Luke Harris, Eric Sanders, Tom Warden
Actuarial Case Reserves
The use of case reserves in actuarial development triangles is ubiquitous. Many of the problems encountered in loss reserving stem from systematic changes and inaccuracies in the determination of case reserves. Case reserves currently serve two primary roles – to facilitate the appropriate settlement of each claim, and to provide financial information. These goals are intrinsically at odds with each other.
As a profession, we need to move beyond the use of subjectively determined case reserves to using case reserves that are more appropriate for loss reserving, that we have constructed directly, using objective claim and exposure information. During this session we will discuss how the separation of the dual roles of case reserves will benefit not only the actuaries in their reserving and pricing work, but also the claim settlement function.
Source:
2020 Spring Meeting
Type:
Concurrent Session
Moderators:
Kathryn Walker
Panelists:
Christopher Gross
Parametric insurance - a growing opportunity
This presentation will give an overview of parametric insurance, explain some case examples in the market, and review of how they work.
Source:
2020 Spring Meeting
Type:
Concurrent Session
Moderators:
HongTao Wang
Panelists:
Jonathan Charak, Sebabrata Sarkar
C-4: Banking on Actuarial Talent: Why Actuaries are Well Positioned to Boost the Banking Sector
The banking industry represents fertile ground for actuaries to deploy their analytical skills. Credit Risk Transfer (CRT) transactions have exploded over the last five years, shifting credit risks traditionally held by banks and governmental entities to the insurance and capital markets. And the adoption of Current Expected Credit Losses (CECL) will require banks and other financial institutions to book provisions for long term credit risks on their balance sheets upon loan origination (think IBNR rather than case reserves). This session will dive into the following key factors leading to the demand for actuarial talent in the banking sector:
• CRT and CECL are changing the way banks and insurance companies think about credit risks;
• Thus, advanced modeling and analytical skills are highly sought for credit risk management; and
• Therefore, actuarial talent has much to contribute to the proper measurement and understanding of these risks.
Source:
2019 Annual Meeting
Type:
Concurrent Session
Panelists:
Michael Schmitz
C-42: Reserving with Machine Learning: Innovations from Loyalty Programs to Insurance
How is machine learning used for loyalty programs and what can it teach us about insurance claim reserving?
While triangular methods have been a foundational tool for decades, individual claim reserving gives the actuary far more information about changes and trends in the liability. Yet the commonly used individual claim reserving techniques leave some of the most valuable data unexamined.
In this session, we’ll cover the benefits of reserving at the individual claim level and describe an approach that sits at the intersection of data science and actuarial science. This session will also introduce a new actuarial tool – the snapshot date triangle – and demonstrate how it can be combined with machine learning to produce a robust and powerful individual claim reserving system. You will learn why the snapshot date triangle was originally developed for estimating loyalty program liabilities and how it can be used in insurance contexts.
Source:
2019 Annual Meeting
Type:
Concurrent Session
Panelists:
Len Llaguno, Julie Hagerstrand, Pong Woo
C-16: Driving Under The Influence -- The Insurance of Distracted Driving
Insurance clearly has skin in the ‘Distracted Driving’ game: almost 10% of the over $100 billion annual losses from distracted driving are incurred through insurance. With such a large problem (that’s only becoming larger), we need to figure out how we analyze the problem. This session will be aimed at defining the problem of distracted driving by:
• Defining what is considered distracted driving for insurance purposes
• Giving the legal landscape of distracted driving
• What role telematics plays – data, variables, models
• Pitfalls of analyzing distracted driving data
Source:
2019 Annual Meeting
Type:
Concurrent Session
Panelists:
Kudakwashe Chibanda, David Kidd, Katie DeGraaf
DD-5: Loss Reserve Variability: Two Fundamental Models (Part 3)
Understanding reserve variability has become an important skill for the practicing actuary. Moving from point estimates and deterministic ranges to distributions of possible outcomes is one of the critical quantitative building blocks.
Based on material from the Reserve Variability Limited Attendance Seminar, this mini workshop will cover two commonly used models for reserve variability: Mack and ODP Bootstrap. The instructors will present both the background to the models and how to implement them using hands on exercises.
The mini workshop is limited to the first 40 participants and will be run in two back-to-back concurrent sessions. The first session will introduce the Mack model and the second session will introduce the ODP bootstrap model. Participants will receive Excel files prior to the session and are expected to bring their laptop to the sessions in order to complete the exercises.
Please NOTE: There is pre-reading material required that you will receive upon expression of interest. As there is a capacity of 40, please email Katrina Evans (kevans@casact.org) if you are interested in attending this workshop. As this will be a hands-on session, you MUST bring your own laptop or tablet.
Source:
2019 Annual Meeting
Type:
Deep Dive Workshop
Panelists:
Louise Francis, Mark Shapland
DD-4: Loss Reserve Variability: Two Fundamental Models (Part 2)
Understanding reserve variability has become an important skill for the practicing actuary. Moving from point estimates and deterministic ranges to distributions of possible outcomes is one of the critical quantitative building blocks.
Based on material from the Reserve Variability Limited Attendance Seminar, this mini workshop will cover two commonly used models for reserve variability: Mack and ODP Bootstrap. The instructors will present both the background to the models and how to implement them using hands on exercises.
The mini workshop is limited to the first 40 participants and will be run in two back-to-back concurrent sessions. The first session will introduce the Mack model and the second session will introduce the ODP bootstrap model. Participants will receive Excel files prior to the session and are expected to bring their laptop to the sessions in order to complete the exercises.
Please NOTE: There is pre-reading material required that you will receive upon expression of interest. As there is a capacity of 40, please email Katrina Evans (kevans@casact.org) if you are interested in attending this workshop. As this will be a hands-on session, you MUST bring your own laptop or tablet.
Source:
2019 Annual Meeting
Type:
Deep Dive Workshop
Panelists:
Louise Francis, Mark Shapland
C-7: An Application of Machine Learning in Rating
Modern rating products are generally based on Generalized Linear Models (GLMs); however, under certain circumstances (e.g., low data volume, complex variable interactions, etc.), GLMs fail to provide adequate predictive accuracy. Machine Learning algorithms can produce significant gains in model predictive power but have generally been met with skepticism by the Insurance Industry – a phenomenon that we believe can be primarily attributed to Machine Learning’s reputation as uninterpretable (i.e., they are seen as “black-boxes”) and a lack of conceptual understanding within the industry.
Caolan Kovach-Orr of Verisk will discuss one type of Machine Learning algorithm, gradient boosted trees and show their advantages over Generalized Linear Models.
Lijuan Zhang of AIG will discuss how to use the output of models to build a rating structure.
Source:
2019 Annual Meeting
Type:
Concurrent Session
Panelists:
Lijuan Zhang, Caolan Kovach-Orr
DD-2: Cyber Insurance Overview (Part 2)
At the 2018 Annual CAS Meeting in Las Vegas, I asked attendees to my cyber panel sessions "What is the biggest challenge that still faces the cyber insurance industry?" There were 227 respondents and 65% of them replied "Unknown unknowns." One year later, we will revisit this question and find out whether the cyber insurance market has become more "known" or if the unknowns continue to evoke unease.
The presenter will provide an overview of the market as well as an update on key trends. Major challenges currently facing the industry as well as potential solutions will also be discussed. There will be ample opportunities for audience participation including live poll questions as well as pauses for Q&A during and after the presentation.
Please note: There is pre-reading material recommended:
The Aon NAIC study which provides an overview of the cyber market:
http://thoughtleadership.aon.com/Documents/201906-us-cyber-market-update.pdf
The Zurich/Advisen survey which looks at cyber insurance buyer trends:
https://www.advisenltd.com/2019-information-security-and-cyber-risk-management-zurich-cyber-survey/
The attached Guy Carpenter / CyberCube study “Looking Beyond the Clouds” on cyber catastrophe losses.
Source:
2019 Annual Meeting
Type:
Deep Dive Workshop
Panelists:
Steven White, Eduard Alpin, Jonathan Laux, Danny Arnett, Michelle Chia
C-19: Extended Warranty Insurance for Electric Vehicles
As electric vehicles (EV) start to gain more and more attention in recent years, EV battery safety and reliability have become a major concern of consumers. The speaker will address the current China EV market situation, extended warranty product design and other related topics.
Source:
2019 Annual Meeting
Type:
Concurrent Session
Panelists:
Ran Guo, Ran Kan
C-27: Ineffective ERM – Facing Obstacles to Achieving Full Potential- A Case Study
Join is in an interactive audience participation session where the panelists will illustrate and “act-out” a real life but disguised case study that considers obstacles to achieving full potential in ERM Programs such as incomplete buy-in from internal stakeholders, behavioral biases, incentive misalignment, board misunderstanding, and incomplete integration of ERM into decision making.
The Speakers will lead an interactive discussion on ways to correct these suboptimal elements within the ERM framework using this case study
Source:
2019 Annual Meeting
Type:
Concurrent Session
Panelists:
Robert Wolf, Michael Speedling
C-18: Enterprise Risk Management – Professional Standards of Practice
Actuaries working in Enterprise Risk Management have a variety of standards and tools to help guide them. The speakers will provide an interactive illustration of a variety of ERM concepts with the audience, such as risk, risk limit, risk appetite, risk tolerance, risk profile, risk mitigation, and economic capital.
The session will include background, case studies, and games to test and apply your knowledge of ASOP46 , ASOP 47 and the new ASOP 55 on Capital Adequacy effective November 1, 2019
Source:
2019 Annual Meeting
Type:
Concurrent Session
Panelists:
Robert Wolf, Michael Speedling
C-46: Systemic Risk Quantification and Implication for Carrier Solvency
One professor from a university in China presents recent research.
(Prof. Zhang had to cancel: Speaker: Zhang Lin (Linda), Hunan University
Title: SVM-SRISK Method)
Professor: Yang Lin, XiAnjiaoTong Liverpool University
Title: Surplus stability analysis and control strategy in general insurance
Abstract: In the insurance industry, actuaries and management team experience significant challenges for pricing a competitive, but also fair premium and keeping accurate level of reserve which lead inevitably to numerous adjustments over time, and potentially several millions of US dollars in annual losses. As model and parameter uncertainty play key roles for actuaries, decision and policy makers, the implementation of advanced mathematical and statistical techniques is highly required. Over the last two decades, applications of regime switching models to finance and economics have received strong attention, particularly among market practitioners. This research attempts to consider how a linear regime switching system in discrete-time framework could be applied to calculate the medium- and long- term accumulated reserves (surplus) and the relevant premiums strategy from the point of view of a non-life insurer. In this direction, we explore the stability of surplus process and insurer's control strategy over a long time period.
Source:
2019 Annual Meeting
Type:
Concurrent Session
Panelists:
Ran Guo, Lin Yang
R-3: Real World meets Industry Standards: IFoA/CAS Paper tested by Marketplace Realities
• A joint IFoA-CAS Working Party, featuring a cross-section of actuaries, underwriters, and academics, produced a research paper to analyze data and information gaps related to pricing global property per risk coverages and the competitive marketplace. Results from surveys of members in the UK, European and US actuarial and other communities, indicated a clear disconnect between the desired information, and the information commonly available for pricing.
• The resulting paper fills this global literary void, as well as presenting a broad range of related pricing topics, including various real world behavioral economics aspects. The paper won two prestigious awards: the IFoA/GIRO 2016 UK Brian Hey award, and the CAS 2019 Hachemeister award. Much of this information is also appropriate for usage in other property and casualty lines of business.
• This session will present an overview of the paper, including a more detailed review of some of the key chapters and real world aspects of this reference document. However, while much has been written about e.g. the growing importance of data scientists, data algorithms, and the role of AI, it seems that the originally identified information gap still exists and in fact in some areas getting larger with less adequate information being provided.
Source:
2019 Annual Meeting
Type:
Concurrent Session
Panelists:
Ana Mata, John Buchanan
R-2: A New and Flexible Regression Model for Ratemaking and Reserving
Understanding the effect of policyholders' risk profile on the number and the amount of claims, as well as the impact of the dependence among different types of claims, is critical to insurance ratemaking and IBNR-type reserving. To accurately quantify the aforementioned features, it is essential to develop a regression model which is flexible, interpretable and statistically tractable for those purposes.
Our actuarial research group at the University of Toronto has recently developed a highly flexible nonlinear regression model, called the logit-weighted reduced mixture of experts (LRMoE) model, for fitting multivariate claim frequencies or severity distributions. The LRMoE model is interpretable as it has two components: Gating functions to classify policyholders into various latent sub-classes and Expert functions to govern the distributional properties of the claims. Because of its flexibility the model can fit any type of claim data accurately and hence minimizes the issue of model selection. An efficient parameter estimation procedure with R codes is developed to fit the model to data.
Model implementation is illustrated using a real automobile insurance data set from a major European insurance company. We first fit the multivariate claim frequencies form the dataset using a counting expert function. Apart from showing excellent fitting results, we are able to interpret the fitted model in an insurance perspective and to visualize the relationship between policyholders' information and their risk level. We demonstrate how the fitted model may be useful for insurance ratemaking.
The second illustration deals with insurance loss severity data that often exhibits heavy-tail behavior, complex distributional characteristics such as multimodality and peculiar links between policyholders' risk profile and claim amounts. Using a Transformed Gamma as the expert function, our model is applied to fit the severity and reporting delay components of the dataset. Apart from obtaining excellent goodness-of-fit, the proposed model is also shown to be useful and crucial for an adequate prediction of the IBNR reserves through an out-of-sample testing.
Source:
2019 Annual Meeting
Type:
Concurrent Session
Panelists:
Andrei Badescu
C-47: Takaful through the Eyes of an Investor: Opportunities and Challenges
Takaful refers to an insurance model that is compliant with Islamic principles - it is about risk sharing (similar to a cooperative or a mutual insurance) rather than risk transfer (i.e. the business model for a stock insurance company). The global takaful market reached USD19 billion in 2017, and is projected to double in size by 2023. This growth prospect provides an opportunity for investors to tap into the young and rising affluent Muslim population (Muslims make up one fifth of the total global population, but only 1% to 2% in North America). Note that takaful is just another way of doing insurance; it is not limited to Muslims only.
This presentation provides an introduction to takaful, its differences and similarities compared to insurance and mutual models, the challenges and opportunities in takaful as a business investment, as well as the role than an actuary can play within takaful.
Source:
2019 Annual Meeting
Type:
Concurrent Session
Panelists:
Nurul (Syuhada) Nurazmi
C-51: University Actuarial Programs and the Property and Casualty Industry – CAS University Award Winners Share Their Insights and Best Practices
Interested in learning how companies can partner with universities to prepare the next generation of property and casualty actuaries? Want to know the innovative ways universities are incorporating property and casualty into their curriculum, research, and industry engagement initiatives? Then attend this session presented by representatives from the 2019 CAS University Award winning schools: Illinois State University, University of Connecticut, and University of Toronto.
Source:
2019 Annual Meeting
Type:
Concurrent Session
Panelists:
Patricia Teufel, Krzysztof Ostaszewski, Samuel Broverman
C-14: Demographics will challenge Japan’s Insurers and Banks
Both insurers and banks are facing challenges from Japan’s declining and aging population. The demographic changes will be particularly detrimental to small regional banks, while large banks and insurers can better mitigate the impact.
In this session, we will illustrate how Japanese P&C insurers are addressing the challenges associated with these demographic changes, as well as with the decline in car ownership among young people.
We then compare the impact of demographic changes on P&C insurers with the impact on Japan’s life insurance and banking sectors.
Source:
2019 Annual Meeting
Type:
Concurrent Session
Panelists:
Soichiro Makimoto, Tomoya Suzuki, Brian Suzuki
C-36: Participation Reverse Mortgages - AND - The Path to Credential in China
Two professors from universities in China present recent research.
Speaker: Pu Ming, SWUFE
Title: Reverse Mortgage Design under House Price Uncertainty
Abstract: In this paper, we propose a novel financial product which can convert the home equity of elder homeowners into continuous cash flow while protecting the lender from the excessive loss from the systematic risk of the real estate market. Motivated by the idea of Amortizing Participation Mortgage (APM), we come up with Participation Reverse Mortgage which allows the lender to lower annuity payment level in the situation where the house price drops below a prefixed level. This product can provide reasonable protect for the lender from the loss of the systematic shock of the real estate market to enhance the supply of the reverse mortgage products. Finally, we provide numerical examples for a typical lender and compared her risk with that for conventional reverse mortgage by comparing the VaRs. We also perform the sensitivity analysis to market parameters such as house price growth rate, volatility and loan interest rate.
Speaker: Zhang LianZeng
Title: The Challenges of Actuarial Education in China
Abstract: Actuaries in North America follows a well developed credential path with curricula designed to match their needs in the job market. In the developing world, especially China, the environment is quite different.
Source:
2019 Annual Meeting
Type:
Concurrent Session
Panelists:
Ran Guo, Lianzeng Zhang, Ming Pu
C-6: Application of Big Data in Insurance
Two professors from universities in China present recent research.
Speaker: Zheng Sujin, CUFE
Title: Is It a Disservice to Apply Social Media in the Forewarning of Natural Disaster?
Abstract: Social network is playing an increasingly important role as an early warning system, which will aid the rapid assessment of disaster and post-disaster reconstruction. In this paper, we seek to answer the following questions: What kind of information combined with data mined from social network will lead to a more accurate result in terms of damage estimation?
Speaker: Gao Guangyuan, Renmin University
Title: Evaluation of driving risk at different speeds
Abstract: Telematics car driving data describes drivers’ driving characteristics. This research studies the driving characteristics at different speeds and their predictive power for claims frequency modeling. We first extract covariates from telematics car driving data using K-medoids clustering and principal components analysis. These telematics covariates are then used as explanatory variables for claims frequency modeling, in which we analyze their predictive power. Moreover, we use these telematics covariates to challenge the classical covariates usually used in practice.
Source:
2019 Annual Meeting
Type:
Concurrent Session
Panelists:
Guangyuan Gao, SuJin Zheng, Feng Chen
