IFoA SA10 study guide
IFoA SA10 · Work through your exam, one topic at a time.
SA10 Data Science is a Specialist Advanced subject in the Institute and Faculty of Actuaries (IFoA) qualification pathway. The IFoA describes itself as the professional body whose curriculum pages hold "All the information you need under our current curriculum", with exams pages holding "Key info for exams including dates and essential resources". SA10 sits within the Specialist Advanced stage, which candidates typically reach after the Core Principles and Core Practice subjects, and it is aimed at actuaries who want to apply data science techniques to insurance, pensions and wider financial work rather than treat modelling as a purely technical exercise. Study should combine statistical and machine learning theory with applied judgement: model selection and validation, data quality and governance, communicating uncertainty to non-technical stakeholders, and the professional and ethical questions that arise when models influence decisions about individuals. The IFoA SA10 study guide has 4 sections and about 1,286 words.
01Data Science: Data Lifecycle, Quality and Reproducibility
This domain examines whether you can take a dataset from source to defensible analysis: framing the business question, assessing data quality, preparing data without leaking information, and making the work reproducible. The data lifecycle follows CRISP-DM: business understanding, data understanding, data preparation, modelling, evaluation, deployment.
341 words, about 2 min02Data Science Applications in Actuarial Work
This domain tests whether you can place data science techniques in an actuarial business context and recognise where regulation, model risk and operational constraints change the answer. Pricing and underwriting: risk classification, telematics, usage-based insurance and quote conversion.
331 words, about 2 min03Statistical and Machine Learning Methods
This domain tests method selection, validation design and honest reporting of performance. GLMs are interpretable, with link functions and distributional assumptions; regularisation such as ridge, lasso and elastic net handles many predictors.
328 words, about 2 min04Professional and Ethical Considerations in Data Science
This domain tests professional judgement when data use creates ethical, legal and reputational risk. The IFoA Actuaries' Code requires integrity, competence and care, impartiality, compliance and clear communication.
286 words, about 1 min