Free SOA Exam FAM (Fundamentals of Actuarial Mathematics) Introduction to Credibility Practice Questions

Master introductory credibility concepts for Exam FAM. Questions test limited fluctuation credibility, greatest accuracy credibility, and Bühlmann credibility models.

40 Questions
27 Easy
7 Medium
6 Hard
2026 Syllabus

Sample Questions

Question 1 Easy
If the full credibility standard for frequency is 1,083, what is the full credibility standard for aggregate losses when severity has a coefficient of variation of 1.5?
Solution
n0agg=n0freq×(1+CV2)=1,083×(1+1.52)=1,083×(1+2.25)=1,083×3.25=3,519.753,520n_0^{\text{agg}} = n_0^{\text{freq}} \times (1 + \text{CV}^2) = 1{,}083 \times (1 + 1.5^2) = 1{,}083 \times (1 + 2.25) = 1{,}083 \times 3.25 = 3{,}519.75 \approx 3{,}520.

(A) is 1.5×n01.5 \times n_0. (B) is 2×n02 \times n_0 (CV = 1). (C) is 2.5×n02.5 \times n_0. (D) is 5×n05 \times n_0 (CV = 2).
Question 2 Medium
Which of the following is NOT a property of the credibility factor ZZ in limited fluctuation credibility?
Solution
In limited fluctuation credibility, ZZ depends only on the sample size relative to the full credibility standard n0n_0. It does NOT depend on the prior distribution. The prior distribution is relevant in Buhlmann (greatest accuracy) credibility, not limited fluctuation.

(A) True: ZZ is always between 0 and 1. (B) True: Z=n/n0Z = \sqrt{n/n_0} increases with nn. (C) True: Z=1Z = 1 when nn0n \ge n_0. (E) True by definition.
Question 3 Hard
An insurer classifies risks into two equally likely classes:

- **Class 1:** Claims follow a Poisson distribution with mean 3
- **Class 2:** Claims follow a Poisson distribution with mean 7

A risk is selected at random and observed for 4 periods, producing a sample mean of 6 claims per period.

Using Bühlmann credibility, calculate the expected number of claims for this risk in the next period.
Solution
B is correct.

For Bühlmann credibility, we need the process variance vv, the variance of hypothetical means aa, and k=v/ak = v/a.

Since each class is equally likely:
v=E[Var(XΘ)]=12(3)+12(7)=5v = E[\text{Var}(X|\Theta)] = \frac{1}{2}(3) + \frac{1}{2}(7) = 5
(For Poisson, variance = mean.)

μ=E[E(XΘ)]=12(3)+12(7)=5\mu = E[E(X|\Theta)] = \frac{1}{2}(3) + \frac{1}{2}(7) = 5

a=Var[E(XΘ)]=12(35)2+12(75)2=12(4)+12(4)=4a = \text{Var}[E(X|\Theta)] = \frac{1}{2}(3-5)^2 + \frac{1}{2}(7-5)^2 = \frac{1}{2}(4) + \frac{1}{2}(4) = 4

k=va=54=1.25k = \frac{v}{a} = \frac{5}{4} = 1.25

With n=4n = 4 observations:
Z=nn+k=44+1.25=45.25=1621=0.7619Z = \frac{n}{n+k} = \frac{4}{4+1.25} = \frac{4}{5.25} = \frac{16}{21} = 0.7619

Bühlmann estimate:
μ^=ZXˉ+(1Z)μ=0.7619×6+0.2381×5=4.5714+1.1905=5.7619\hat{\mu} = Z \bar{X} + (1-Z)\mu = 0.7619 \times 6 + 0.2381 \times 5 = 4.5714 + 1.1905 = 5.7619

Rounded: 5.76.
Choice A is incorrect because 5.00 equals the overall mean μ\mu and assigns zero credibility to the data.
Choice E is incorrect because 6.00 equals the sample mean Xˉ\bar{X} and assigns full credibility to the data.
Choice C is incorrect because 5.50 is the simple average of μ\mu and Xˉ\bar{X}, which implies Z=0.50Z = 0.50.
Choice D is incorrect because 6.24 results from computing k=a/v=0.80k = a/v = 0.80 (inverting the ratio), giving Z=4/4.80=0.833Z = 4/4.80 = 0.833.
Create a Free Account to Access All 40 Questions →

More Exam FAM Topics

About FreeFellow

FreeFellow is a free exam prep platform for actuarial (SOA & CAS), CFA, CFP, CPA, CAIA, and securities licensing candidates. Every question includes a detailed solution. Full lessons, flashcards with spaced repetition, timed mock exams, performance analytics, and a personalized study plan are all included — no paywalls, no ads.