Limited Attendance Seminars
Limited attendance seminars are designed to be interactive in nature, allowing participants to get more hands-on-training from the instructors in various topics relating to the actuarial profession. Attendance is limited to a maximum of 35-40 students depending on the seminar. Attendees will be selected on a first registered, first accepted basis. Sometimes participants are expected to bring their own laptop to the seminar.
No Limited Attendance Seminars are being offered at this time.
INTRODUCTION TO R LAS
Typically held in December
The Introduction to R Limited Attendance Seminar focuses on the practical issues involved in using R in typical types of actuarial analysis. The focus of the seminar will be familiarize the student with R, introduce packages related to common actuarial work, and demonstrating how to integrate R with spreadsheets.
Attendees will benefit from training in the following areas:
- Foundations of R (Installing, Commands, Data Exploration and Exchange)
- Topics in Reserving
- Topics in Predictive Modeling
- Topics in Loss Distributions and Statistics
- Getting the most out of the R community
Participants are expected to bring their own laptop to the seminar. Instructions will be sent prior to the seminar on installing R and related packages. Some minor preliminary work will be required in order to maximize our time during the seminar.
PREDICTIVE MODELING LAS
Typically held twice a year – Spring/Fall
The Predictive Modeling Limited Attendance Seminar focuses on the practical issues involved in analyzing insurance data and building predictive models. Attendees will benefit from training in the following areas:
- Data scrubbing and manipulation
- Data Exploration, variable selection, and modeling methodologies
- Model validation considerations
- Interpreting predictive model results
The seminar will impart a systematic methodology for developing modeling solutions for property and casualty insurance business problems.
The curriculum will focus on critical issues, and highlight common pitfalls, of developing such solutions. The seminar will provide a review of relevant statistical concepts and Generalized Linear Model [GLM] theory. However, the focus of the seminar will be on practical applications, and theoretical concepts will be conveyed in the midst of a sequence of data analysis case studies.
The seminar is designed to be accessible to actuaries with little or no predictive modeling experience. However, actuaries with predictive modeling experience have taken the seminar in past years and reported satisfaction with it. Note also that the predictive modeling workshop held at the 2013 CAS RPM seminar covered a subset of the material to be covered during the Limited Attendance Seminar.
The seminar is designed to be interactive in nature, enabling participants to manipulate and explore data, and build models and analyze models in real time along with the instructors. To this end, the seminar will include a self-contained introduction to the R statistical computing environment. R is a sophisticated and powerful modeling tool that has rapidly gained attention both within and outside the actuarial profession. For more on R, see Professor Jed Frees' web site on installing and using R.
In addition to a review of basic theory and expository case studies, the seminar will present three business applications of predictive modeling:
- Factor optimization for an existing personal lines auto class plan using Generalized Linear Models to model pure premium
- Development of a personal auto "scorecard" model and integration with the underlying rating plan
- Development of a commercial lines pricing/underwriting algorithm.
RESERVE VARIABILITY LAS
Typically held once every two years
This seminar is designed to enhance the skills of the practicing actuary with regard to fitting and using and communicating results from loss reserve models. Emphasis in the seminar will be on the process of moving from deterministic methods for estimating a single point to stochastic models for estimating a distribution.
The learning objectives include:
- Review of Statistical Concepts
- Understanding of Ranges vs. Distributions
- Knowledge of Statistical Modeling Techniques
- Hands on Use of Models, with Emphasis on Simulation Models
- Understanding of Diagnostic Testing
- Understanding of Model Strengths & Weaknesses
- A Better Understanding of Quantifying and Communicating Uncertainty