ACT 210 · Year 2 · Semester 2 · 2 credits · Integration & Practice

R & Python for Actuaries

Bridging classical actuarial mathematics and modern predictive engineering through bilingual, reproducible computation in R and Python.

Full practice exam · 30 Q · 120 min

Season 1 · 8 episodes

The Actuarial Codebase

When an eight-million-dollar spreadsheet error threatens Northbridge Mutual with regulatory seizure, second-year actuarial analyst Maya Lin must rebuild the insurer's entire computational core across R and Python into an auditable, reproducible production engine.

Protagonist · Maya Lin, an ambitious second-year actuarial analyst at Northbridge Mutual with dual training in loss modeling and software engineering.

Setting · Northbridge Mutual's high-pressure actuarial department in Chicago, navigating state regulatory audits, legacy mainframes, and modern polyglot data stacks.

Stakes · A looming state insurance department audit threatens to revoke Northbridge's commercial underwriting license unless every reserve, pricing model, and life valuation is proved fully reproducible under actuarial standards of practice.

Pick an episode

Episodes

Season finale: Maya deploys an end-to-end automated polyglot pipeline that flawlessly replicates Thomas Mack's 1993 benchmark, reconciles the multi-million dollar discrepancy, and establishes the company's first fully governed actuarial codebase.

From the lecturer: Welcome to ACT 210. In this course, you will transform the mathematical equations from life contingencies, loss models, and reserving into living, production-ready software. To get the most out of these lecture films, set up a split-screen workspace: the lecture on one side and an active IDE (RStudio or VS Code with an active R and Python terminal) on the other. Keep an engineering notebook beside your keyboard. Before running any package command—whether it is MackChainLadder in R or a Poisson GLM in statsmodels—sketch the underlying matrix dimensions, loss triangle coordinates, or life table lookup logic by hand. Never copy and paste code passively. Type every line, predict the output object structure before hitting execute, and inspect the memory footprint. While this course does not target a single written paper, it provides the computational engine for the advanced SOA (ALTAM, ASTAM, PA) and CAS (Exams 5, 8) syllabi, instilling the professional coding standards and ASOP 56 model governance principles required in modern actuarial practice.

Cold open

At seven forty-five on a rainy Monday morning, Maya Lin opens Northbridge's master commercial auto workbook. Cell F-one-hundred-and-forty-two displays an eight-point-four million dollar reserve deficit, caused entirely by an intern dragging an unanchored VLOOKUP formula across forty-eight accident quarters.

preparing lecture…

Transcript

Structure tabular insurance data according to tidy data principles using tibbles
Manipulate and transform policyholder and claims records using core dplyr verbs
Reshape development triangles and multi-year rating tables using tidyr pivot functions
Apply vectorised operations to compute actuarial metrics such as loss ratios and pure premiums efficiently