Because of its statistical and graphical capabilities, R has become part of the actuarial toolkit. However, actuaries often lack systematic training in manipulating data using R.
In this session, we review fundamental R data structures and their application to insurance data. Participants will learn to perform basic actuarial procedures using modern data manipulation libraries (tidyverse).
We will cover different aspects of a reserving and ratemaking workflow (policy/claim data manipulation, rating, loss development, indications), with examples taken from CAS Exam 5 study notes (Werner, Appendix C). In each module, we will discuss practical R programming problems in small groups.
Participants will gain a better grasp of R data structures and their application to insurance data. They will be able to script basic data manipulation pipelines that perform routine actuarial procedures. Participant should have basic programming skills, as well as some experience performing actuarial procedures.
Learning Objectives:
Select the most appropriate R data structure to represent different types of actuarial concepts: vector, list, data frame, array, date, etc.
Write R scripts that load, manipulate and export insurance data, using state-of-the-art libraries: tidyverse, dplyr, lubridate, purrr, etc.
Perform simple standard actuarial procedures in R: aggregate claim/policy data, calculate summary ratios, calculate and develop claim triangles, etc.
