MAS-II Linear Mixed Models
MAS-II mixed models add hierarchy to regression: fixed effects describe population-level effects, random effects describe group-level variation, and shrinkage controls noisy group estimates.
CAS Exam MAS-II
Current MAS-II page and 2026 outline are mapped for format, domain weights, assumed knowledge, tables, and the reading list that drives credibility, mixed models, statistical learning, and time series preparation.
What the official PDFs establish
- Appointment length
- 4.5-hour appointment with a 4-hour exam duration.
- Scheduled break
- The appointment includes a scheduled 15-minute break plus tutorial/confidentiality/survey time.
- Assumed knowledge
- Calculus, probability, linear algebra concepts at the regression-prerequisite level, and mastery of MAS-I concepts are assumed.
Topic and domain coverage
| Topic | Weight | Source |
|---|---|---|
| Introduction to Credibility | 15-25% | |
| Linear Mixed Models | 10-20% | |
| Statistical Learning | 40-50% | |
| Time Series with Constant Variance | 15-25% | |
| Cognitive level: Remember | 5-10% | |
| Cognitive level: Understand and Apply | 55-60% | |
| Cognitive level: Analyze and Evaluate | 35-40% | |
| Cognitive level: Create | 0-5% |
Chapter and reading intelligence
- Tse
Credibility work is assigned from Nonlife Actuarial Models, covering classical, Buhlmann, Buhlmann-Straub, and Bayesian credibility sections in chapters 6-9.
- West
Linear Mixed Models: A Practical Guide Using Statistical Software is assigned across all chapters, excluding coding examples, with shrinkage notes called out separately.
- James et al., Salis, and GLM Monograph
Statistical learning is anchored to ISLR chapters 2.2, 4.4.2, 8, 10, and 12, Salis chapters 3 and 10, and Chapter 7 of Generalized Linear Models for Insurance Rating.
- Cowpertwait and Metcalfe
Time series preparation uses Introductory Time Series with R chapters 1-5 excluding selected sections, plus chapter 6 and sections 7.1-7.3.
Official files used by the map
- CAS content outlinecontent-outline
Primary source for domain weights, exam format, assumed knowledge, and official reading assignments.
Source: MAS-II Content Outline 2026
Quick Answer
The current MAS-II outline gives linear mixed models a 10-20% weight and assigns mixed-model reading from West. The domain is smaller than statistical learning, but it is easy to lose points if fixed effects, random effects, and variance components blur together.
The actuarial use case is natural: territories, classes, companies, accident years, or policyholder groups can vary around a population-level pattern.
Fixed And Random Effects
Fixed effects estimate population-level relationships. Random effects model group-level deviations from those relationships. A mixed model uses both at once.
If a model has territory random effects, the estimate for a small territory will usually shrink more strongly toward the overall mean than the estimate for a large territory. That is the same statistical instinct candidates see in credibility.
Variance Components
Variance components tell you how much variation lives between groups and how much remains at the observation level. MAS-II can ask for interpretation, not only identification.
A large group-level variance means groups differ materially after fixed effects are accounted for. A small group-level variance means the grouping may add little beyond the fixed-effect structure.
Output Reading
Mixed-model output usually separates fixed-effect estimates, random-effect variance components, fit criteria, and residual diagnostics. Read those sections separately before making a recommendation.
Do not interpret a random effect as if it were a regular fixed-effect coefficient for every observation. It is an estimated group deviation, and its precision depends on group information and model assumptions.
Mixed-model mistakes
| Mistake | Fix |
|---|---|
| Calling every group factor a fixed effect because it appears in the data. | Ask whether the model estimates one coefficient per named group or models group deviations as random. |
| Ignoring variance components after reading fixed effects. | Use variance components to explain how much grouping still matters. |
| Treating shrinkage as an arbitrary penalty. | Connect shrinkage to group exposure, noise, and between-group variation. |
Original Practice Drill
A claim severity model has fixed effects for vehicle age and coverage limit, with random intercepts by territory. Explain what the random intercept is estimating and why a territory with low exposure may have an estimate closer to the overall intercept.
A complete answer connects the random intercept to territory-level deviation and explains shrinkage toward the population mean when group information is limited.