IAI Actuarial Core Principles · Actuarial Statistics · Expectations and conditional expectations
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(N)?
Var(N) is 10/3. By the conditional variance formula, Var(N) = E[Λ] + Var(Λ) for a Poisson with random mean. Here E[Λ] = 2 and Var(Λ) = 4/3, so the total is 2 + 4/3 = 10/3.
- A2
- B4/3
- C10/3Correct
- D3
- 14/3
Explanation
E[Λ] = 2 and Var(Λ) = 16/12 = 4/3. Var(N) = E[Var(N|Λ)] + Var(E[N|Λ]) = E[Λ] + Var(Λ) = 2 + 4/3 = 10/3. Option 4/3 ignores the process variance part; option 2 ignores the variance of Λ.
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)?
- X and Y have joint density f(x,y) = x + y on 0<x<1, 0<y<1. What is E[Y | X = 1/2]?
- X is uniform on (0, 6). What is E[X²]?
- A non-negative integer-valued random variable N has P(N > n) = 0.6^n for n = 0, 1, 2, .... What is E[N]?
- Given Λ, X is Poisson with mean Λ. Λ has mean 3 and variance 2. What is Var(X)?
- A random variable N takes the values 0, 1, 2, ... with P(N = n) = (1/2)^(n+1). What is E[N]?