IAI Actuarial Core Principles · Paper CS1
IAI Actuarial Statistics (CS1) Paper Guide
Actuarial Statistics (CS1) is the IAI Core Principles subject on data analysis, random variables, inference, regression and Bayesian methods. You solve it by learning each method's formula and assumptions, practising written answers with full working, and using R for Paper B. Paper A is written. Paper B is computer-based.
CS1 Actuarial Statistics tests whether you can use probability and statistics to model data and draw conclusions. It sits in the Actuarial Statistics module of Core Principles. The 2026 syllabus weights are: Data analysis 10%, Random variables and distributions 20%, Statistical inference 25%, Regression theory and applications 30%, and Bayesian statistics 15%. Regression and inference together carry the most weight, so they deserve the most of your time.
The subject has two examinations. Paper A is a written exam of 3 hours 15 minutes, and recent Paper A question papers carry 100 marks. Paper B is a 1 hour 45 minute computer-based exam, where you work with data in R. Check the latest IAI notice for the exact format of your session, because the number of multiple-choice questions and the structure can change between sessions.
From the November 2025 session, the pass rule for CS1 is at least 30% in each of Paper A and Paper B and 50% in aggregate, with Paper A and Paper B weighted 70:30. IAI may amend this session by session and confirms it with results. In practice, students lose marks in two ways. They know the theory but skip the working, or they know the theory but cannot run and interpret code under time pressure. Because Paper A carries the larger weight, a strong written paper does most of the work. Paper B still needs steady practice, as you must clear its minimum.
Actuarial Statistics: chapters and topics
Data analysis
Purpose and function of data analysis
Data analysis
Exploratory data analysis
Random variables and distributions
Basic univariate distributions and generating samples
- Discrete Distributions: Binomial, Poisson, Geometric
- Continuous Distributions: Normal, Lognormal, Exponential, Gamma
- Uniform Distribution and Distribution Properties
- Generating Random Samples: Inverse Transform Method
- Acceptance-Rejection and Other Simulation Methods
- Simulation Accuracy and Number of Simulations
Random variables and distributions
Jointly distributed random variables
Random variables and distributions
Expectations and conditional expectations
Random variables and distributions
Generating functions
Random variables and distributions
Central limit theorem
Random variables and distributions
Random sampling and sampling distributions
Statistical inference
Estimators and their properties
Statistical inference
Confidence intervals and prediction intervals
Statistical inference
Hypothesis testing and goodness of fit
Regression theory and applications
Linear regression models
Regression theory and applications
Generalised linear models
Bayesian statistics
Bayesian inference: priors, posteriors, loss functions and credible intervals
Bayesian statistics
Credibility theory: Bayesian and empirical Bayes credibility premiums
How to prepare Actuarial Statistics
Plan for steady work across the whole syllabus, with extra time on regression and inference. Build theory first, then switch to timed questions and R practice. The steps below follow the order of the chapters.
- Map the syllabus to the five topic areas and their weights. Give regression (30%) and inference (25%) the largest share of your hours, and still cover every area.
- Start with the foundations: data analysis, exploratory methods, univariate distributions, joint distributions, expectations and conditional expectations. Write each distribution's mean, variance and key properties on one revision sheet, and learn to derive them, not just recall them.
- Learn generating functions and the central limit theorem as tools. Practise finding moments from a generating function and using the CLT approximation, and state its conditions each time.
- Work through inference in sequence: sampling distributions, estimators and their properties, confidence and prediction intervals, then hypothesis testing and goodness of fit. For every method, write the assumptions, the test statistic, the distribution under the null and the conclusion in words.
- Study linear regression and generalised linear models together. Be able to set up the model, estimate parameters, interpret coefficients, check residuals and explain why a GLM is chosen over a linear model.
- Cover Bayesian inference and credibility theory. Practise finding posteriors, choosing a loss function, building credible intervals, and calculating Bayesian and empirical Bayes credibility premiums.
- Learn R alongside the theory, not after it. For each chapter, run the matching analysis on a small data set, read the output and write one sentence of interpretation. Practise in a timed setting before Paper B.
- In the last weeks, do full past papers under exam conditions. Mark yourself strictly on working and wording, list every lost mark by cause, and revise those causes first.
Time management in the exam
- Spend the first few minutes reading the whole paper. Pick out the multiple-choice questions and the written questions you can start at once.
- Answer multiple-choice questions quickly. Each is worth only 2 marks, so do not let one hold you up. Mark the hard ones and return if time remains.
- Use the marks as a clock. A question worth about 10 marks should not take more than about a tenth of the writing time, plus a short margin for checking.
- If you are stuck on a part, write down the formula and setup to pick up method marks, then move on. Later parts often accept your earlier answer.
- In Paper B, check your data import and variable names early. Small errors in setup can cost many minutes later. Save and label your outputs as you go.
- Keep the last few minutes for checking units, signs, degrees of freedom and whether each conclusion answers the question asked.
Mistakes that cost marks in Actuarial Statistics
Memorising formulas without stating assumptions
Fix: Before using any method, write its assumptions in one line, such as normality, independence or known variance. Examiners reward this.
Giving a number with no interpretation
Fix: End every inference answer with a sentence in context: what the interval, p-value or estimate means for the problem.
Confusing similar ideas, such as confidence intervals, prediction intervals and credible intervals
Fix: Keep a short comparison note. Say what is random, what is fixed and what the interval covers in each case.
Neglecting Paper B until late in the preparation
Fix: Practise R weekly from the start. Remember Paper B has its own minimum mark, so weak code can sink an otherwise good result.
Weak regression and GLM understanding despite its high weight
Fix: Spend your largest block of time here. Practise deriving estimates, reading output, checking residuals and justifying model choice.
Skipping working in written answers
Fix: Show the formula, substitution and result on separate lines. A wrong final value can still earn most of the marks if the method is clear.
Actuarial Statistics: frequently asked questions
What does CS1 Actuarial Statistics cover?
It covers data analysis, random variables and distributions, statistical inference, regression theory and applications, and Bayesian statistics. The 2026 weights are 10%, 20%, 25%, 30% and 15% respectively. Credibility theory is part of the Bayesian area in this guide's chapter list.
How is CS1 examined?
There are two examinations. Paper A is written and lasts 3 hours 15 minutes. Paper B is a 1 hour 45 minute computer-based exam. Always confirm the current format in the latest IAI notice, as details can change by session.
What is the pass mark for CS1?
From the November 2025 session, you need at least 30% in each of Paper A and Paper B and 50% in aggregate. Paper A and Paper B are weighted 70:30. IAI may amend this session by session and confirms it with results.
Which topics should I prioritise?
Give most time to regression and statistical inference, as they carry the highest weights. Do not skip the others, because every area can appear in both papers. Revise weak areas using timed past-paper questions.
Do I need R for CS1?
Yes. Paper B is computer-based and tests practical analysis with R. Practise running models, reading output and writing short interpretations well before the exam.