Event Details
Deloitte LLC
111 S Wacker Dr.
Chicago, IL 60606
November 6, 2017
About This Event
The Introductory Predictive Modeling Limited Attendance Seminar [PMLAS-1] focuses on the practical issues involved in analyzing insurance data and building predictive models. Only basic recall of the statistical concepts features on actuarial exams will be presupposed, and significant time will be spent on "ground up, first principles" discussions of Ordinary Least Squares Regression and Generalized Linear Models.
Registration for this event is now closed.
Registration for the Intermediate Predictive Modeling LAS must be done separately.
The Introductory Predictive Modeling Limited Attendance Seminar [PMLAS-1] focuses on the practical issues involved in analyzing insurance data and building predictive models. Only basic recall of the statistical concepts features on actuarial exams will be presupposed, and significant time will be spent on "ground up, first principles" discussions of Ordinary Least Squares Regression and Generalized Linear Models. More experienced practitioners with working knowledge of Generalized Linear Models and the R statistical computing environment might wish to skip this seminar in favor of the Intermediate seminar [PMLAS-2].
Attendees of PMLAS-1 will benefit from training in the following areas:
- Data scrubbing and manipulation
- Data exploration, variable selection, and modeling methodologies
- Interpreting predictive model results
- Model validation and economic benefit calculation
The seminar will impart a systematic methodology for developing modeling solutions for property and casualty insurance business problems. The focus on will be practical, emphasizing critical issues, common pitfalls, and the real-world context of developing such solutions. The philosophy behind the seminar's coverage of statistical first principles is captured by the adage, "Nothing is more practical than a good theory." (Kurt Lewin). The theoretical emphasis will be on conceptual understanding to facilitate more effective model building, model criticism, and model refinement.
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 one of the most comprehensive and widely used statistical modeling tools. For more on R, see Professor Jed Frees' web site on installing and using R. (Note some seminar participants elect to observe and take notes rather than participate in the hands-on data analysis. This is a perfectly reasonable way to approach the seminar.)
The various techniques and methodologies will be illustrated through a variety interactive case studies, ranging from small "textbook" case studies illustrating specific points to more realistic case studies illustrating modeling methodology and the "art" of data analysis.
Content Covered:
- Statistics refresher: maximum Likelihood, simulating data, bootstrapping, simple linear regression
- Generalized Linear Models (Gaussian, Poisson, Gamma, logistic, Tweedie)
- Spline regression and Generalized Additive Models
- Cross-validation and the Bias-Variance Tradeoff
- Methodological concepts such as Exploratory Data Analysis, model design, nested model comparisons, model criticism, missing data, and variable selection will be discussed and illustrated throughout the seminar
- Case study involving models designed for point estimation versus ranking cases. This case study will compare and contrast the construction of rating plans and underwriting models. Practical design considerations depending on the intended use include:
- Modeling data structure and data processing
- Target variable selection and model design
- Structure of predictive variables and predictive variable development
- Model development methodology
- Model validation scenarios and approaches
- Considerations on business implementation
Attendance is limited to a maximum of 35 students. Attendees will be selected on a first-registered, first-accepted basis. Participants are expected to bring their own laptop, loaded with the R software, to the seminar. Instructions on how to install R and perform basic manipulations in R will be provided in advance of the seminar.
CONTINUING EDUCATION CREDIT
The CAS Continuing Education Policy applies to all ACAS and FCAS members who provide actuarial services. Actuarial services are defined in the CAS Code of Professional Conduct as "professional services provided to a Principal by an individual acting in the capacity of an actuary. Such services include the rendering of advice, recommendations, findings, or opinions based upon actuarial considerations." Members who are or could be subject to the continuing education requirements of a national actuarial organization can meet the requirements of the CAS Continuing Education Policy by satisfying the continuing education requirements established by a national actuarial organization recognized by the Policy.
Participants should claim credit commensurate with the extent of their participation in the activity. CAS members earn 1 CE Credit per 50 minutes of educational session time, not to include breaks or lunch.
Note: The amount of CE credit that can be earned for participating in this activity must be assessed by the individual attendee. It also may be different for individuals who are subject to the requirements of organizations other than the American Academy of Actuaries.
CANCELLATIONS
Registrations fees will be refunded for cancellations received in writing at the CAS Office via fax, 703-276-3108, or email, refund@casact.org, by November 22, 2017 less a $200 processing fee.
CONTACT INFORMATION
- For more information on content, please contact Jody Allen, Professional Education and Research Coordinator, at jallen@casact.org.
- For more information on attendee registration, please email the Actuaries' Resource Center at arc@casact.org.
- For more information on the Seminar other than registration or content issues, please email meetings@casact.org.
- For more information on other CAS opportunities or regarding administrative policies such as complaints and refunds, please contact the CAS Office at (703) 276-3100 or office@casact.org.
REGISTRATION FEES
Registration for this event is now closed.
Registration for the Intermediate Predictive Modeling LAS must be done separately.
REGISTRATION FEES (IN U.S. DOLLARS) |
IF RECEIVED |
IF RECEIVED |
CAS Member, iCAS Member, or Active Candidate* |
$975 |
$1,175 |
Non-Members |
$1,175 |
$1,375 |
* An Active Candidate is a non-CAS member who has attempted at least one actuarial exam in the last two years.
LOCATION AND LODGING
Location:
Deloitte LLC
111 S Wacker Dr.
Chicago, IL 60606
A room block has not been designated for this meeting. The following hotels are near the Deloitte office:
Hyatt Place Chicago Downtown
28 North Franklin Street
Phone: (312) 955-0950
La Quinta Inn and Suites Downtown
1 South Franklin Street
Phone: (312) 558-1020
JW Marriott Chicago
151 West Adams Street
Phone: (312) 660-8200
Hyatt Centric the Loop Chicago
100 West Monroe Street
Phone: (312) 236-1234
W Chicago City Center
172 West Adams Street
Phone: (312) 332-1200
INSTRUCTORS
Jim Guszcza is the US Chief Data Scientist of Deloitte Consulting, and a member of Deloitte's Advanced Analytics and Modeling practice. He has built and helped design predictive models both in insurance and a variety of other public and private sector domains. Jim has also served as an assistant professor of Actuarial Science, Risk Management, and Insurance at the University of Wisconsin-Madison. A frequent contributor to actuarial seminars and publications, Jim has co-taught the Casualty Actuarial Society's Limited Attendance Seminar in Predictive Modeling each year since 2006. Jim is a Fellow of the Casualty Actuarial Society and currently serves on its Board of Directors. He is a graduate of St. John's College in Santa Fe, New Mexico and has a PhD in Philosophy from the University of Chicago.
Jun Yan is a leading modeler in Deloitte Consulting's Actuarial, Risk, and Analytics practice. Since joining Deloitte, he has designed and developed numerous P&C predictive models for both personal and commercial lines applications. Jun is a frequent speaker in CAS seminars and SAS data mining conferences. Jun has co-taught the Casualty Actuarial Society's Limited Attendance Seminar each year since 2006. Jun has published several papers in CAS Actuarial Forum. Prior to joining Deloitte, Jun was a senior consultant in personal line research at The Hartford Insurance Group (HIG). At HIG, Jun developed class plan for personal auto and homeowners, credit scoring, growth and profitability projection, and claim level loss development. Jun holds a Ph.D. in Statistics from Indiana University in Bloomington, Indiana.
SCHEDULE
December 5, 2017 8:30 a.m. - 5:30 p.m.
December 6, 2017 8:30 a.m. - 12:00 p.m.
Note: Schedule is subject to change