Skip to main content

Mathematical Statistics I · Bias, risk, consistency, and point estimation

The unbiased sample variance is not the MSE winner

Problem

Mathematical Statistics I practice problem

Let X1,…,XnX_1,\ldots,X_n be independent N(μ,σ2)N(\mu,\sigma^2) with both parameters unknown and n≥2n\ge2. Write Q=∑i=1n(Xi−Xˉ)2Q=\sum_{i=1}^n(X_i-\bar X)^2 and S2=Q/(n−1)S^2=Q/(n-1). Under squared-error loss for the target σ2\sigma^2, compute the exact risk of S2S^2 and of the MLE σ^2=Q/n\hat\sigma^2=Q/n. Then, for a constant a≥0a\ge0, compute the exact risk of aS2aS^2 and find the unique a∗a^* minimizing that risk. Identify the corresponding estimator as a multiple of QQ and compute its risk.

Prove thatR(σ2,S2)>R(σ2,σ^2)>R(σ2,a∗S2)R(\sigma^2,S^2) >R(\sigma^2,\hat\sigma^2) >R(\sigma^2,a^*S^2)for every σ2>0\sigma^2>0. This is a comparison of estimators of a scale parameter. Do not compare linear estimators aXaX of a normal mean, and do not invoke a decision-theoretic shrinkage argument for μ\mu.

Hint

Hint

Unbiasedness of S2S^2 makes its risk equal to its variance.

Check your work

Work the problem yourself first. Then open it in Training to check your answer and read the full worked solution.

The answer check and full solution for this problem come with ProofAnvil Practice membership ($19 USD monthly). See membership. Or start with the free Mathematical Statistics I sample problem: Try the free sample problem.

More Mathematical Statistics I practice problems