IAI Actuarial Core Principles · Actuarial Statistics · Expectations and conditional expectations
Let N be Poisson with mean 3, and given N = n, S is the sum of n independent claims each with mean 100 and variance 400 (S = 0 if n = 0). What is Var(S)?
Var(S) is 31,200. By the law of total variance, it is 3×400 for claim variability plus 3×100² for variation in claim number, giving 1,200 + 30,000.
- A30,000
- B31,200Correct
- C40,000
- D1,200
- 9,000
Explanation
Var(S) = E[Var(S|N)] + Var(E[S|N]) = E[N]×400 + Var(N)×100² = 3×400 + 3×10,000 = 1,200 + 30,000 = 31,200. Option 30,000 omits the within-N variance term; 1,200 omits the variation in N.
Did you get it right without looking?
One question tells you little. A timed set on Expectations and conditional expectations shows your real accuracy, how long you take and where you lose marks.
More Expectations and conditional expectations questions
- Given Λ = λ, the claim count N is Poisson(λ). Λ has an exponential distribution with mean 2. What is Var(N)?
- The number of claims N on a policy in a year has a Poisson distribution with mean Λ, where Λ is uniformly distributed on (0, 4). What is Var…
- A non-negative integer-valued random variable N has P(N > n) = 0.6^n for n = 0, 1, 2, .... What is E[N]?
- A random variable X has E[X] = 4 and Var(X) = 9. Let Y = 3 − 2X. What is E[Y²]?
- A claim amount X (in rupees thousand) has density f(x) = 2/x^3 for x > 1 and 0 otherwise. What is E[X]?
- X is uniformly distributed on (0, 6). What is E[X²]?