Agnostic tests can control the type I and type II errors simultaneously

DOI10.1214/19-BJPS431zbMATH Open1445.62009arXiv1805.04620MaRDI QIDQ783299FDOQ783299

Authors: Victor Coscrato, Rafael Izbicki, Rafael B. Stern 

Publication date: 12 August 2020

Published in: Brazilian Journal of Probability and Statistics (Search for Journal in Brave)

Abstract: Despite its common practice, statistical hypothesis testing presents challenges in interpretation. For instance, in the standard frequentist framework there is no control of the type II error. As a result, the non-rejection of the null hypothesis cannot reasonably be interpreted as its acceptance. We propose that this dilemma can be overcome by using agnostic hypothesis tests, since they can control the type I and II errors simultaneously. In order to make this idea operational, we show how to obtain agnostic hypothesis in typical models. For instance, we show how to build (unbiased) uniformly most powerful agnostic tests and how to obtain agnostic tests from standard p-values. Also, we present conditions such that the above tests can be made logically coherent. Finally, we present examples of consistent agnostic hypothesis tests.

Full work available at URL: https://arxiv.org/abs/1805.04620

Recommendations

• scientific article; zbMATH DE number 804002
• Rejection odds and rejection ratios: a proposal for statistical practice in testing hypotheses
• A note on type S/M errors in hypothesis testing
• Adaptative significance levels using optimal decision rules: balancing by weighting the error probabilities
• scientific article; zbMATH DE number 1337355

zbMATH Keywords

hypothesis testlogical consistencyuniformly most powerful teststhree-decision problem

Mathematics Subject Classification ID

Parametric hypothesis testing (62F03) Foundations and philosophical topics in statistics (62A01) Compound decision problems in statistical decision theory (62C25)

Cites Work

• Testing Statistical Hypotheses
• Bayes Factors
• Title not available (Why is that?)
• Title not available (Why is that?)
• Classification with confidence
• Mathematical Statistics
• Symmetry, invariance and ontology in physics and statistics
• Modes of Parametric Statistical Inference
• Tests of statistical hypotheses and their use in studies of natural phenomena
• Continuous versions of Haack’s puzzles: equilibria, eigen-states and ontologies
• Logical consistency in simultaneous statistical test procedures
• Logically-consistent hypothesis testing and the hexagon of oppositions
• Least Ambiguous Set-Valued Classifiers With Bounded Error Levels
• A primer on linear models.

Uses Software

• BayesFactor

This page was built for publication: Agnostic tests can control the type I and type II errors simultaneously

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q783299)