Higher-order approximate confidence intervals

DOI10.1016/J.JSPI.2020.11.013zbMATH Open1465.62086arXiv1811.11031OpenAlexW3110984560MaRDI QIDQ830725FDOQ830725

Authors: Eliane C. Pinheiro, Silvia L. P. Ferrari, Francisco M. C. Medeiros 

Publication date: 7 May 2021

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

Abstract: Standard confidence intervals employed in applied statistical analysis are usually based on asymptotic approximations. Such approximations can be considerably inaccurate in small and moderate sized samples. We derive accurate confidence intervals based on higher-order approximate quantiles of the score function. The coverage approximation error is $O(n^{-3/2})$ while the approximation error of confidence intervals based on the asymptotic normality of MLEs is $O(n^{-1/2})$. Monte Carlo simulations confirm the theoretical findings. An implementation for regression models and real data applications are provided.

Full work available at URL: https://arxiv.org/abs/1811.11031

Recommendations

• Higher-order comparisons of asymptotic confidence intervals
• JACKKNIFING: HIGHER ORDER ACCURATE CONFIDENCE INTERVALS
• Higher-order approximations for interval estimation in binomial settings
• More accurate confidence intervals in exponential families
• scientific article; zbMATH DE number 806840
• Confidence intervals from stochastic approximation
• Approximate and generalized approaches to confidence intervals on a variance component

zbMATH Keywords

nuisance parametersregression modelsmaximum likelihood estimatesaccurate confidence intervalsmodified score equations

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) Nonparametric tolerance and confidence regions (62G15) Applications of statistics to social sciences (62P25)

Cites Work

• Bootstrap confidence intervals. With comments and a rejoinder by the authors
• Bias reduction of maximum likelihood estimates
• Median bias reduction of maximum likelihood estimates
• Beta Regression for Modelling Rates and Proportions
• A GENERAL METHOD FOR APPROXIMATING TO THE DISTRIBUTION OF LIKELIHOOD RATIO CRITERIA
• Title not available (Why is that?)
• Likelihood methods in statistics
• Title not available (Why is that?)
• Title not available (Why is that?)
• Bias prevention of maximum likelihood estimates for scalar skew normal and skew \(t\) distribu\-tions
• The likelihood ratio criterion and the asymptotic expansion of its distribution
• A modified score test statistic having chi-squared distribution to order \(n^{-1}\)
• Title not available (Why is that?)
• Log-symmetric distributions: statistical properties and parameter estimation
• An introduction to Bartlett correction and bias reduction
• APPROXIMATE CONFIDENCE INTERVALS
• Likelihood asymptotics
• Tensor notation and cumulants of polynomials
• On bootstrap and analytical bias corrections
• Improved Score Tests in Symmetric Linear Regression Models
• Routes to higher-order accuracy in parametric inference
• Tensors and likelihood expansions in the presence of nuisance parameters
• Gradient statistic: higher-order asymptotics and Bartlett-type correction
• A semiparametric approach for joint modeling of median and skewness
• Modern Likelihood‐Frequentist Inference
• Graphical methods for investigating the finite-sample properties of confidence regions

Cited In (9)

• A note on likelihood asymptotics in normal linear regression
• On the Construction of Bayes–Confidence Regions
• Title not available (Why is that?)
• Uniform coverage bounds for confidence intervals and Berry-Esseen theorems for Edgeworth expansion
• Confidence intervals for the mean value of response function in generalized linear models
• Higher order accurate confidence intervals for smooth functions of means with application to the correlation coefficient
• Confidence distributions from likelihoods by median bias correction
• Bounds for the sample size to justify normal approximation of the confidence level
• Higher-order comparisons of asymptotic confidence intervals

Uses Software

• R
• betareg
• Ox
• likelihoodAsy

This page was built for publication: Higher-order approximate confidence intervals

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q830725)