Note on a paradox in decision-theoretic interval estimation

DOI10.1016/J.SPL.2012.08.021MaRDI QIDQ1933707zbMATH OpenOpenAlexFDO

Authors Paul Kabaila

Publication date 25 January 2013

Published in Statistics \& Probability Letters (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1205.5074

zbMATH Keywords

Bayes' ruleconfidence intervaldecision theoryinterval estimator

Mathematics Subject Classification ID

Bayesian problems; characterization of Bayes procedures (62C10) Admissibility in statistical decision theory (62C15) Parametric tolerance and confidence regions (62F25)

Abstract: Confidence intervals are assessed according to two criteria, namely expected length and coverage probability. In an attempt to apply the decision-theoretic method to finding a good confidence interval, a loss function that is a linear combination of the interval length and the indicator function that the interval includes the parameter of interest has been proposed. We consider the particular case that the parameter of interest is the normal mean, when the variance is unknown. Casella, Hwang and Robert, Statistica Sinica, 1993, have shown that this loss function, combined with the standard noninformative prior, leads to a generalized Bayes rule that is a confidence interval for this parameter which has "paradoxical behaviour". We show that a simple modification of this loss function, combined with the same prior, leads to a generalized Bayes rule that is the usual confidence interval i.e. the "paradoxical behaviour" is removed.

Recommendations

• scientific article; zbMATH DE number 774831
• Interval estimation of normal means with respect to asymmetric loss functions
• Generalized Confidence Intervals
• scientific article; zbMATH DE number 775730
• Brown's paradox in the estimated confidence approach

Cited in

(4)

• scientific article; zbMATH DE number 3984317 (Why is no real title available?)
• scientific article; zbMATH DE number 774831 (Why is no real title available?)
• A “Paradox” in Confidence Interval Construction Using Sufficient Statistics
• Choosing a coverage probability for forecasting the incidence of cancer

This page was built for publication: Note on a paradox in decision-theoretic interval estimation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1933707)