MAS-II Credibility
MAS-II credibility is the shrinkage domain: use individual experience when it is credible, pull toward a collective mean when it is noisy, and explain what the weight means.
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 credibility a 15-25% weight. The reading map includes classical credibility, Buhlmann credibility, Buhlmann-Straub credibility, and Bayesian credibility from nonlife actuarial models.
The exam wants more than formula recall. A candidate has to say what is being blended, what variance component controls the weight, and why the resulting estimate is credible or not credible in context.
Core Idea
Credibility estimates balance a risk's own experience against a broader portfolio estimate. High credibility means the risk's experience has enough exposure or low enough noise to carry more weight.
Low credibility does not mean the risk data are useless. It means the estimate should shrink toward the collective mean because the observed experience is too noisy to stand alone.
Buhlmann-Straub Role
Buhlmann-Straub extends the basic idea to unequal exposures. A claim cost based on more exposure is usually more precise than one based on less exposure, so the weighted mean replaces the simple mean.
For MAS-II, do not treat the exposure weights as decoration. They change the risk mean estimate, the credibility weight, and the interpretation of how much information the risk has supplied.
Bayesian Link
Bayesian credibility makes the shrinkage logic explicit through prior and posterior thinking. MAS-II candidates should be able to translate between actuarial credibility language and statistical updating language.
The practical question is the same in both dialects: how much should the new data move the estimate away from the prior or collective view?
Original Practice Drill
A territory has observed loss cost 1,250 from exposure 40. The collective mean is 1,000, EPV is 900, and VHM is 30. Compute the credibility estimate using Buhlmann-Straub weight shape Z = w / (w + K), then explain whether the result is closer to the territory or collective mean.
A complete answer computes K = 30, Z = 40 / 70, and an estimate near 1,143. It also says the estimate moves toward the territory experience but remains pulled toward the collective mean.