Inadmissibility of the uncombined two-stage estimator when additional samples are available

DOI10.1007/BF00053065MaRDI QIDQ1117618zbMATH OpenOpenAlexFDO

Authors Tatsuya Kubokawa

Publication date 1988

Published in Annals of the Institute of Statistical Mathematics (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/bf00053065

zbMATH Keywords

additional samplescombined estimatorsmean of a normal distributionpreassigned bound on the risktwo-stage estimator

Mathematics Subject Classification ID

Admissibility in statistical decision theory (62C15) Point estimation (62F10)

Review text

Consider the problem of constructing an estimator with a preassigned bound on the risk for a mean of a normal distribution. The paper shows that the usual two-stage estimator is improved on by combined estimators when additional samples taken from distributions with the same mean and different variances are available.

Recommendations

• Improving on two-stage estimators for scale families
• On utilizing information from a second sample in estimating variance
• A note on admissibility when precision is unbounded
• On two-stage james-stein sstimators
• Inadmissible estimators of normal quantiles and two-sample problems with additional information

Cites work

• A note on estimating the common mean of k normal distributions and the stein problem
• Admissible minimax estimation of a common mean of two normal populations
• Estimating common parameters of growth curve models
• Estimation of a common mean and recovery of interblock information
• Estimation of a common mean of two normal distributions
• Estimation of location parameters from two linear models under normality
• scientific article; zbMATH DE number 3945039 (Why is no real title available?)
• Linear Statistical Inference and its Applications
• On two-stage james-stein sstimators
• Point and confidence estimation of a common mean and recovery of interblock information
• Sequential shrinkage estimation
• Sequential shrinkage estimation of linear regression parameters
• Two inequalities with an application

Cited in

(2)

• Improving on two-stage estimators for scale families
• Inadmissible estimators of normal quantiles and two-sample problems with additional information

This page was built for publication: Inadmissibility of the uncombined two-stage estimator when additional samples are available

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1117618)