IAI Actuarial Core Principles · Actuarial Statistics · Estimators and their properties
X1,...,Xn are independent N(μ, σ²) with μ known to be 0. The estimator S = (1/n)ΣXi² is used for σ². Another analyst uses the estimator V = (1/(n+1))ΣXi². Which statement is correct?
S is unbiased and V has bias -σ²/(n+1). With a known zero mean each Xi² has expectation σ², so S is unbiased. V divides the same sum by n+1, giving expectation nσ²/(n+1), which is below σ² by σ²/(n+1).
- AS is biased and V is unbiased
- BBoth are unbiased
- CS is unbiased and V has bias -σ²/(n+1)Correct
- DS is unbiased and V has bias +σ²/(n+1)
- Both are biased, with S having bias -σ²/n
Explanation
E[Xi²] = σ² since μ = 0, so E[S] = σ² and S is unbiased. E[V] = nσ²/(n+1), so bias = nσ²/(n+1) − σ² = −σ²/(n+1). V underestimates; the sign is negative.
Did you get it right without looking?
One question tells you little. A timed set on Estimators and their properties shows your real accuracy, how long you take and where you lose marks.
More Estimators and their properties questions
- X1,...,Xn is a random sample from a uniform distribution on (0, θ). Let M = max(Xi). Which statement is correct?
- For a sample of size n from a Poisson distribution with mean λ, the maximum likelihood estimate of P(X = 0) is required. With n = 50 and a s…
- A random sample of size n is taken from a Normal(μ, σ²) distribution. Consider the estimator S*² = (1/n)Σ(Xi - X̄)² of σ². Which statement i…
- A sample of n independent observations is taken from a uniform distribution on (0, θ). The estimator θ̂ = 2X̄ is used for θ. What is its mea…
- A claims analyst models the number of claims per policy as Poisson with mean λ. Over 50 policies the total number of claims is 85. What is t…
- X1,...,Xn are independent Poisson(λ) variables. An analyst estimates λ² by T = (X̄)². Which expression gives the bias of T?