Event Details
LAUNCH ACADEMY
77 SUMMER ST, 7TH FLOOR
BOSTON, MASSACHUSETTS
July 24, 2017
About This Event
This is a four day-long course that will give the attendee an introduction to R, with the objective of preparing attendees to be able to apply what they've learned upon completion of the course. To that end, the focus is on direct, interactive content delivery.
This is a four day-long course that will give the attendee an introduction to R, with the objective of preparing attendees to be able to apply what they've learned upon completion of the course. To that end, the focus is on direct, interactive content delivery.
Attendance is limited to 30 participants. Attendees are expected to bring their own laptop, which they will use for Pre-Work and the Bootcamp.
LEARNING OBJECTIVES
Attendees will leave the course with a working knowledge of R, having applied it in a realistic setting. More specifically, participants will exit the course having mastered the following:
- Reading data into the R environment from a variety of sources
- Visualizing data and using this as the basis of selecting a model
- Fitting several different models to a set of data, assessing the quality of fit and choosing between competing models
- Manipulating data so that it may be modeled, or otherwise communicated
- Using the "actuar" and "ChainLadder" packages, which are designed for actuarial applications
CAPSTONE PROJECTS
The CAS RBootcamp will provide general instruction in a classroom environment on how to use R. The bootcamp will also provide capstone projects in three different actuarial areas for applying classroom learning objectives. Attendees will share project outcomes with the whole group on the last day using tools available in the broader “R Environment.” Attendees must choose their preferred topic during the registration process.
Ratemaking
Bootcamp attendees who sign up for this track will be given datasets representing premium and claims for a product line. The project will be to use R to analyze the data for trends and anomalies and identify and fit a model to the data that distinguishes between predictive characteristics. Track attendees will prepare disruption analyses, profitability projections, and other management information to communicate the effect of the new model on the portfolio.
Reserving
Attendees who sign up for this track will be given a dataset that contains loss triangles for several different lines of business in several different territories. Participants will conduct a reserve study using standard techniques like chain ladder as well as less common methods like the Clark growth curve model and hierarchical regression.
Reinsurance
Bootcamp attendees who sign up for this track will be given datasets representing premium and claims for an insurance company. The project will be to use R to choose among a set of covers offered by the market. Track attendees will prepare appropriate metrics for consideration and other management information to communicate the recommendation.
CONTACT INFORMATION
- For more information on content, please contact Nora Potter, Education Coordinator, at npotter@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
This event is now sold out. Please complete the form to be added to the event wait list.
REGISTRATION FEES (IN U.S. DOLLARS) | IF RECEIVED ON/BEFORE JULY 24, 2017 | IF RECEIVED |
CAS Member, or Active Candidate* | $2,800 | $2,900 |
Non-Members | $3,000 | $3,100 |
The CAS Institute Members | $2,800 | $2,900 |
* An Active Candidate is a non-CAS member who has attempted at least one actuarial exam in the last two years.
CANCELLATION POLICY
 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 August 1, 2017 less a $100 processing fee.
LODGING INFORMATION
This event is not located at a hotel, so attendees will need to make their own hotel reservations. Nearby hotels include:
Hyatt Regency Boston
The Godfrey Hotel Boston
The Ritz-Carlton, Boston
Club Quarters Hotel in Boston
W Boston
INSTRUCTORS
Brian Fannin is the Captain at PirateGrunt LLC, an actuarial, data and predictive modelling consultancy. Brian has held a number of positions at both primary and excess insurance companies, both in the US and overseas. He is the chair of the CAS open source software committee, with a focus on R. His principle areas of research are in stochastic reserving, predictive modelling and visualization of data. Brian believes that objective analysis of data will lead to the delivery of social justice.
Adam L. Rich is a Pricing Actuary at Beazley Group in Farmington, CT and directs the Specialty Lines Analytics group. Adam has been programming computers for over twenty years and has been using and promoting the use of R for the past six.
Daniel M. Murphy, FCAS, MAAA, heads an independent actuarial consulting practice, Trinostics LLC, in the San Francisco Bay Area. Dan has been employed at various P&C companies and consulting firms in California, holding the Chief Actuary position at Argonaut Insurance in the 1990’s. He is an enthusiastic proponent of the R open-source environment for implementing actuarial algorithms, most notably a contributor to the ChainLadder package on CRAN. Dan has taught math and statistics courses at local colleges and enjoys research in the area of stochastic modeling of loss reserves, earning the Woodward-Fondiller award in 1994 for his CAS Proceedings paper “Unbiased Loss Development Factors.” After graduating with a BS in Mathematics from Santa Clara University and an MS in Statistics from the University of Illinois, he began his working career in operations research and financial programming with the Southern Pacific railroad.
TOPIC SCHEDULE
Download PDF of Tentative Schedule (may have minor changes closer to the bootcamp)
PRE-WORK
- Install R, RStudio
- Read the raw book
- A few basic exercises
DAY 1: GETTING COMFORTABLE WITH THE ENVIRONMENT, PRELIMINARY PROGRAMMING
AM
- Advanced Environment, RStudio setup
- Basics review and more
- Saving and sourcing scripts
- Installing and using packages
- Data structures
- Vectors, lists
- Working with data frames
- Reading/writing data
- Reading to/from CSV, hadley's readr and readxl packages
- Reading to/from sqlite with RODBC or DBI
- Basic data analysis in R
- Linear regression
- Fit a distribution
- Histograms/kernel density plots
- Basic visualization
- PM
- AM review
- Advanced visualization with ggplot2
- Introduction to capstone projects
- Overview of online resources: StackOverflow, r-bloggers
DAY 2: RMARKDOWN, DPLYR, BEGIN CAPSTONE WORK
- AM
- Day 1 review/questions
- Augment R scripts with RMarkdown
- dplyr to select, summarize and transform data
- PM
- Begin capstone work
DAY 3: GLMS AND OTHER PREDICTIVE MODELS
- AM
- Day 2 review/questions
- GLMs
- Decision trees
- PM
- Assisted work on capstone
DAY 4: FURTHER PRACTICE, CAPSTONE
- AM
- Complete capstone
- Generate actuarial report document in Word or PDF (or HTML)
- PM
- Complete actuarial report
- Present to larger group
- (if time) Sample presentations from other R groups around the country
