Skip to main content

Mathematical Statistics I · Sufficiency, completeness, and exponential families

A complete exponential sum yields the survival UMVUE

Problem

Mathematical Statistics I practice problem

Let n≥2n\ge2 and let X1,…,XnX_1,\ldots,X_n be independent exponential random variables with rate θ>0\theta>0, so that each has density θe−θx\theta e^{-\theta x} on (0,∞)(0,\infty). Write T=∑i=1nXiT=\sum_{i=1}^n X_i. Show that the model is a one-parameter exponential family and that TT is a complete sufficient statistic. Identify the law of TT.

The target is τ(θ)=Pθ(X1>1)=e−θ\tau(\theta)=P_\theta(X_1>1)=e^{-\theta}. Prove that δ=1{X1>1}\delta=\mathbf 1\{X_1>1\} is unbiased for τ(θ)\tau(\theta). Then computeδ∗(T)=E[δ∣T]\delta^*(T)=\mathbb E[\delta\mid T]explicitly, including the case T≤1T\le1, and verify that the displayed formula does not contain θ\theta. Conclude that δ∗\delta^* is the UMVUE of e−θe^{-\theta}. Also give the UMVUE of 1/θ1/\theta.

Do not replace this calculation by a Poisson zero-count Rao–Blackwellization, and do not treat the problem as a Neyman–Pearson test of a point null.

Hint

Hint

The joint density is θnexp⁡(−θT)\theta^n\exp(-\theta T) on the positive orthant.

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