2018 Ratemaking, Product and Modeling Seminar Program Guide
IET-1: Actionable Insights: How to Make Actuarial Research Come to Life
- by David Novich, Verisk Analytics and Brian Ostrowski, Verisk Analytics
Actuaries provide critical research for the insurance industry. But often, that research is not easily understood by other insurance professionals. In this session, we’ll explore how to transform your findings into actionable insights that are concise, compelling, and easy to understand. Using best practices and case studies, we’ll show you how to describe and illustrate your technical research to a broader audience.
W-6: Advanced Predictive Modeling
*A laptop is required for this workshop and there will be a maximum of 40 participants.
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 workshop will discuss a variety of statistical learning methods that either refine or serve as complements to the GLM framework.
Methods for capturing potentially complex non-linear relationships, integrating credibility, tree-based modeling, and machine learning techniques will all be covered in this workshop. Core themes such as the bias-variance tradeoff and cross-validation will be woven throughout the presentation. This workshop will also include a section on feature engineering from unstructured text data that will focus on using natural language processing tools in R to extract predictive signals from noisy text data.
Attendees will develop a better understanding of more advanced modeling techniques including Generalized Additive Models (GAMs), Hierarchical Models, Penalized Regression (Ridge, Lasso, Elastic Net), CART, random forests, boosting and bagging, and natural language processing tools.
The workshop will assume a working knowledge of GLM modeling and R software.
R-1: Aims and Advances in Personal Lines Risk Classification
Insurance risk classification continues to become more sophisticated and refined, fueled by enhancements in data, analytical techniques and computing power. This session will open with a brief discussion of the aims of risk classification, including how this topic is addressed in actuarial standards of practice and insurance rate regulation and some recent anecdotes of where it’s been covered in the press. We will focus on personal lines insurance and discuss real examples of novel risk classification used in pricing and what issues the classifications aim to address, what data and techniques are used to develop these classifications, and what implementation challenges they may present. We will then examine how telematics, as well as advances in autonomous safety features, has and will continue to change risk classification.
M-4: Analytics Roadmap for Smaller Companies: Investing in the Right Tools and Techniques
- by Drew Lawyer, Earnix and Marcus Deckert, Pekin Insurance
The analytic challenges for a $10 billion company are very different from those of a $100 million company. Smaller companies have always found innovative ways to compete with the larger companies through distribution channels, customer relations, and target marketing. One of the new competitive frontiers in the industry is analytics – companies that are able to effectively leverage analytics are seeing a great return on their investment. Unfortunately for smaller companies, larger carriers are at an advantage simply from the availability of resources.
In this session we will discuss what smaller companies (primarily those under $500 million) can do to compete in the analytics arms race. Not all analytics activities provide the same return on investment; here we will outline some of the most cost effective analytics initiatives spanning across product enhancement, tools, and methodology. This session will also address the talent gap that many smaller companies face and the most effective ways to address it.
M-5: 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, you should understand and demonstrate the goodness of fit of the model. In most cases, you should also prove 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. Key topics include confusion matrices, ROC curves, Gini indices, lift charts, double-lift charts, residual plots, likelihood, penalized likelihood and deviance. Model development is usually a major investment. We should make sure our models are performing well to get the best bang for the buck.
IET-2: Artificial Intelligence and Advanced Insurance Analytics: Techniques, Technology, and Promise
The explosive growth in data science has generated considerable enthusiasm for data analysis across all industries. Exciting new tools developed in the technology industry are driving change across all industries. It is easier than ever to change the technical backbone of a company or department, but it remains difficult for insurers to understand whether they should be pursuing, monitoring, or simply ignoring some of the newest trends. In this discussion, we will recap the basic concepts in artificial intelligence and take a candid approach to filtering through these innovations. We will also discuss a framework for identifying opportunities within your own organization.
RP-1: ASOP on Estimating Future Costs for Prospective Property/Casualty Risk Transfer and Risk Funding
"As part of the traditional ratemaking and other projections of the cost of future fortuitous events, many actuaries are involved in the development of the estimates of the underlying future costs of many risk transfer or risk retention programs – insurance, re-insurance, excess insurance, captives, self-insurance, etc. These estimates serve as the preliminary basis for the development of the filed rates, negotiated contracts or charges to operating costs.
This panel will provide a summary of the recently approved ASOP covering the area of estimating future costs prospectively whether they are transferred or retained.
GS-1: Auto Insurance 2028
- by Matthew Moore, Highway Loss Data Institute, Thomas Karol, National Association of Mutual Insurance Companies and Alexander Timm
With so many changes to auto insurance and the market, how will auto insurance look in the future? A panel of experts from across the industry will discuss the changing market.
W-5: Basic Ratemaking
Part 1: Fundamental Insurance Equations
This session will go over the process for developing a rate indication. Participants will gain familiarity with the steps and adjustments needed. Considerations such as loss development, trend, credibility, etc will be discussed.
Part 2: Ratemaking Relativities
This session will examine the reasons and some methods that actuaries use to allocate overall average rates to various subdivisions of a line of business. Some of the methods discussed will consist of univariate, multivariate, and generalized linear modeling techniques.
Part 3: Estimating Claim Liabilities
The session will focus on estimating claim liabilities including development factor techniques, advantages and disadvantages of different methods, and diagnostics of the various methods
Part 4: Large Account Pricing
This session will examine the differences between large accounts and small accounts, and how they can and should be treated differently from an insurance pricing perspective. Participants will be introduced to pricing methods that both utilize the increased credibility of large account experience and recognize the uniqueness of each risk. Specifically, this session will serve as an introduction to experience rating, schedule rating, and retrospective rating.
IET-3: Blockchain: Finding Real Opportunities Behind the Hype
Blockchain technology is already significantly impacting the insurance industry’s business models and products. Consumer behavior has changed following new needs and expectations. The nature of risk is evolving incorporating new sources of risk and more risk interconnectedness. New ways of measuring and tracking risk in real time allow for innovative risk fragmentation. New players are exploring different ways of allocating risks and interacting with customers. The acceleration in the #insurtech space significantly increased investments in new technologies. One of those new technologies is Blockchain.
This session will navigate the blockchain hype and explore what a blockchain is and isn’t, identify how blockchains can disrupt the insurance industry, and outline the threats and opportunities that players will have to navigate. It will start with a brief Blockchain 101 and then discuss the implications of the technology in more depth by delving into the latest use cases and existing insurance products based on blockchain technology. The last third of this session will be interactive and will provide food for thought to actively discuss the latest blockchain insurance protocols with the audience.