Skip to main content

Introduction to the Use of Machine Learning in the P&C Insurance Industry

Introduction to the Use of Machine Learning in the P&C Insurance Industry

The CAS has hundreds of articles about predictive modeling and about as many hours of live presentations from the field’s leading practitioners. The material ranges from descriptions of the many use case for machine learning in insurance to the basics of the most common algorithms to the practical challenges associated with integrating machine learning in an insurance organization. You could design a sophisticated model to troll through all of that content and generate a rational introduction to machine learning. Or you could let us do that for you.


The CAS has compiled session recordings and research papers on machine learning into one online activity to introduce P&C actuaries to machine learning.

NOTE: This activity might qualify for up to 5.7* Continuing Education credits for the recordings and the actual number of credits to read the papers included as part of the course. Participants should claim credit commensurate with the extent of their viewing and reading. CAS members earn 1 CE credit for every 50 minutes of educational activity. *The amount of CE credit that can be earned for participating in this activity must be assessed by the individual participant. It also may be different for individuals who are subject to the requirements of organizations other than the American Academy of Actuaries.

Objectives:
  1. Introduce the opportunities, benefits and concerns with using machine learning algorithms in the P&C insurance industry
  2. Provide education on the types of machine learning algorithms and how a few representative algorithms work
  3. Recognize and how to overcome potential obstacles to implementation
Modules:
  1. Introduce the opportunities, benefits and concerns with using machine learning algorithms in the P&C insurance industry
    1. Staying Ahead Of The Analytical Competitive Curve: Integrating The Broad Range Applications Of Predictive Modeling In A Competitive Market Environment (Spring 2008 E-Forum)
    2. Comparing Machine Learning And Conventional Statistical Techniques In Claims Models (Recording, 2017 RPM)
    3. A Machine-Learning Approach To Parameter Estimation (2017 Monograph)*
  2. Provide education on the types of machine learning algorithms and how a few representative algorithms work
    1. A Practical Introduction To Machine Learning For Actuaries (Spring 2016 E-Forum)
    2. Distinguishing The Forest From The Trees: A Comparison Of Tree Based Data Mining Methods (Variance)*
    3. Neural Networks Demystified (Winter 2001 E-Forum)*
    4. Easy Tree-Sy: An Overview Of Decision Trees (Recording, 2017 RPM)*
  3. Provide education on the types of machine learning algorithms and how a few representative algorithms work
    1. Using Predictive Modeling For Workers Compensation Ratemaking (Recording, 2016 RPM)*
  4. Q&A Webinar with Select Speakers/Researchers (60 minutes) – TBD
    1. The CAS will send an email notification to all registered users before the webinars to confirm attendance.
* Presenters or authors from these modules have agreed to participate in the Q&A webinars. (If you have questions about any of these items before the Q&A webinar, please email them to dcore@casact.org.)