Finite sample properties of confidence intervals centered on a model averaged estimator

DOI10.1016/J.JSPI.2019.10.004zbMATH Open1437.62118arXiv1906.07933OpenAlexW2952452776WikidataQ126787265 ScholiaQ126787265MaRDI QIDQ2301097FDOQ2301097

Authors: Alan H. Welsh, Christeen Wijethunga, Paul Kabaila 

Publication date: 28 February 2020

Published in: Journal of Statistical Planning and Inference (Search for Journal in Brave)

Abstract: We consider the confidence interval centered on a frequentist model averaged estimator that was proposed by Buckland, Burnham & Augustin (1997). In the context of a simple testbed situation involving two linear regression models, we derive exact expressions for the confidence interval and then for the coverage and scaled expected length of the confidence interval. We use these measures to explore the exact finite sample performance of the Buckland-Burnham-Augustin confidence interval. We also explore the limiting asymptotic case (as the residual degrees of freedom increases) and compare our results for this case to those obtained for the asymptotic coverage of the confidence interval by Hjort & Claeskens (2003).

Full work available at URL: https://arxiv.org/abs/1906.07933

Recommendations

• Model-averaged confidence intervals
• Model-averaged Wald confidence intervals
• Interval estimation by frequentist model averaging
• Exact model averaged tail area confidence intervals
• The performance of model averaged tail area confidence intervals

zbMATH Keywords

coveragemodel averaged confidence intervalscaled expected length

Mathematics Subject Classification ID

Linear regression; mixed models (62J05) Parametric tolerance and confidence regions (62F25)

Cites Work

• Title not available (Why is that?)
• Model Selection: An Integral Part of Inference
• Model Selection and Multimodel Inference
• Model-averaged Wald confidence intervals
• Confidence intervals in regression utilizing prior information
• Upper bounds on the minimum coverage probability of confidence intervals in regression after model selection
• Model Selection and Model Averaging
• Frequentist Model Average Estimators
• On the Large-Sample Minimal Coverage Probability of Confidence Intervals After Model Selection
• Title not available (Why is that?)
• Model-averaged profile likelihood intervals
• Model selection bias and Freedman's paradox
• On the minimum coverage probability of model averaged tail area confidence intervals
• Model-averaged confidence intervals
• The performance of model averaged tail area confidence intervals
• Confidence intervals centred on bootstrap smoothed estimators

Cited In (9)

• Exact model averaged tail area confidence intervals
• Confidence intervals centred on bootstrap smoothed estimators: an impossibility result
• Interval estimation by frequentist model averaging
• Confidence intervals in general regression models that utilize uncertain prior information
• Model-averaged confidence intervals
• Model-averaged profile likelihood intervals
• Model averaged tail area confidence intervals in nested linear regression models
• Model-averaged Wald confidence intervals
• The performance of model averaged tail area confidence intervals

Uses Software

• R
• Glmulti
• AICcmodavg
• MuMIn
• MAMI

This page was built for publication: Finite sample properties of confidence intervals centered on a model averaged estimator

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2301097)