Increasing the replicability for linear models via adaptive significance levels

DOI10.1007/S11749-022-00803-4MaRDI QIDQ2084719zbMATH OpenOpenAlexFDO

Authors Yanyan Li

Publication date 13 October 2022

Published in Test (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2003.06938

zbMATH Keywords

\(p\)-value calibrationadaptive \(\alpha\)Bayes factorlikelihood ratiolinear modelPBIC

Mathematics Subject Classification ID

General considerations in statistical decision theory (62C05) Bayesian problems; characterization of Bayes procedures (62C10) Diagnostics, and linear inference and regression (62J20)

Abstract: We put forward an adaptive alpha (Type I Error) that decreases as the information grows, for hypothesis tests in which nested linear models are compared. A less elaborate adaptation was already presented in citet{PP2014} for comparing general i.i.d. models. In this article we present refined versions to compare nested linear models. This calibration may be interpreted as a Bayes-non-Bayes compromise, of a simple translations of a Bayes Factor on frequentist terms that leads to statistical consistency, and most importantly, it is a step towards statistics that promotes replicable scientific findings.

Recommendations

• Corrected confidence intervals for parameters in adaptive linear models
• Controlling the significance levels of prediction error tests for linear regression models
• Adaptive tests of linear hypotheses by model selection
• Improving linear quantile regression for replicated data
• Prediction of response values in linear regression models from replicated experiments
• scientific article; zbMATH DE number 1361718

Cites work

• Asymptotic expansions for large deviation probabilities of noncentral generalized chi-square distributions
• Calibration of values for testing precise null hypotheses
• Changing statistical significance with the amount of information: the adaptive \(\alpha\) significance level
• Consistency of Bayesian procedures for variable selection
• Counterexamples to parsimony and BIC
• scientific article; zbMATH DE number 46578 (Why is no real title available?)
• scientific article; zbMATH DE number 48701 (Why is no real title available?)
• scientific article; zbMATH DE number 3273551 (Why is no real title available?)
• On the use of non-local prior densities in Bayesian hypothesis tests
• Prior-based Bayesian information criterion
• The ASA Statement on p-Values: Context, Process, and Purpose
• The effective sample size

Cited in

(3)

• Changing statistical significance with the amount of information: the adaptive \(\alpha\) significance level
• Adaptative significance levels using optimal decision rules: balancing by weighting the error probabilities
• Likelihood, Replicability and Robbins' Confidence Sequences

This page was built for publication: Increasing the replicability for linear models via adaptive significance levels

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2084719)