Exam SRM covers statistical learning and predictive modeling fundamentals. Heavy on GLMs and tree-based methods. Typically 100–150 hours of prep. Upload your study notes and PDF handouts and exclam.ai builds adaptive flashcards, quizzes, and a weekly plan tuned to the SRM syllabus.
3.5-hour computer-based test. Multiple-choice questions covering statistical learning concepts. No computer programming is required on SRM — that comes in Exam PA.
No coding on SRM — multiple-choice only. Exam PA is where you actually write R.
GLMs and decision tree-based methods dominate the content (together 60-75%).
Typically one of the higher pass rates among preliminary exams.
Often taken alongside or immediately before Exam PA.
Format details change periodically. Always verify the current exam structure on soa.org before your exam.
The 5 topic areas tested on Exam SRM, with approximate weight bands from the public SOA syllabus. Click any topic for a dedicated study page.
Supervised and unsupervised learning, model assessment, bias-variance tradeoff, and resampling methods including cross-validation.
Ordinary least squares regression, generalized linear models, variable selection, and regularization techniques.
Classification and regression trees, pruning, and ensemble methods including bagging, random forests, and boosting.
Dimension reduction via principal components analysis and unsupervised clustering via k-means and hierarchical clustering.
Introduction to time series analysis including autoregressive, moving average, and ARIMA models.
Topic names and approximate weight bands sourced from the public SOA Exam SRM syllabus on soa.org.
Drop in your ACTEX Exam SRM digital edition (full-corpus PDF), scanned ASM pages, TIA handouts, or your own notes. exclam.ai parses them into structured content for a fully guided study plan.
AI generates flashcards for formulas and concepts, plus multiple-choice quizzes targeting each of the 5 topic areas. Edit any card.
Coverage phase, review phase, and mocks — distributed across your ~12-week timeline or whatever you choose.
Pick the timeline that matches your exam date. exclam.ai spreads roughly 125 hours across coverage, review, and mock phases — each reading paired with its own quiz and FSRS flashcards. Switch the plan length to rebalance the schedule instantly.
A visual preview of how exclam.ai turns your plan into daily reading, flashcard, quiz, review, and mock streaks across the timeline you pick.
Sign in to unlock your full 14 weeks plan
Every module dated to your exam, with quizzes and FSRS flashcards generated from your own materials.
Unlock my full plan freeYour Exam SRM plan walks the syllabus module-by-module before transitioning into review and mock phases. Module weeks are allocated proportional to SOA topic weights — heavier-weighted modules get more time, lighter ones either get a single week or pair with adjacent topics on intensive paces. Change the plan length above to see the schedule rebalance.
Supervised and unsupervised learning, model assessment, bias-variance tradeoff, and resampling methods including cross-validation. Build flashcards covering every learning objective and quiz yourself before moving to the next module.
Ordinary least squares regression, generalized linear models, variable selection, and regularization techniques. Build flashcards covering every learning objective and quiz yourself before moving to the next module.
Classification and regression trees, pruning, and ensemble methods including bagging, random forests, and boosting. Build flashcards covering every learning objective and quiz yourself before moving to the next module.
Dimension reduction via principal components analysis and unsupervised clustering via k-means and hierarchical clustering. Build flashcards covering every learning objective and quiz yourself before moving to the next module.
Introduction to time series analysis including autoregressive, moving average, and ARIMA models. Build flashcards covering every learning objective and quiz yourself before moving to the next module.
Weak-topic drilling. exclam.ai surfaces topics where you underperformed during coverage and re-quizzes them. Daily FSRS flashcard reps across all 5 topics to prevent decay. Start doing timed question batches focused on the heaviest-weight sections.
Full-length Exam SRM practice exams under timed conditions. Target one mock every 3–5 days. Same-day error review: for every question you miss, re-derive the solution from scratch. Taper in the final 3 days — light flashcards only.
exclam.ai works alongside any of these vendors. Upload paths vary, ACTEX sells full digital editions as downloadable PDFs (cleanest upload), ASM print manuals can be scanned for the full corpus, and The Infinite Actuary provides downloadable PDF handouts alongside their videos. For ASM digital (app-locked) and Coaching Actuaries (online platform), upload your own notes taken while studying. We do not distribute or reproduce any vendor content, your manual stays with the publisher.
Yes. SRM covers the conceptual foundations that PA applies in practice with R. Taking SRM first makes PA significantly easier.
No. SRM is entirely multiple-choice and tests your understanding of statistical learning concepts. Exam PA is where R proficiency becomes essential.
It is the most recommended free resource and covers most of the SRM content. Most candidates supplement it with a dedicated study manual and a question bank.
Upload your ACTEX digital edition, scanned ASM pages, TIA handouts, or your own notes, exclam.ai builds a fully guided study plan around your exam date.