Adaptive critical value for constrained likelihood ratio testing

DOI10.1093/BIOMET/ASAA013MaRDI QIDQ5127213zbMATH OpenOpenAlexFDO

Authors Diaa Al Mohamad, Eric A. Cator, Jelle J. Goeman, Erik Van Zwet

Publication date 21 October 2020

Published in Biometrika (Search for Journal in Brave)

Copyright license Creative Commons Attribution-NonCommercial 4.0 International

Full work available at URL https://arxiv.org/abs/1806.01325

zbMATH Keywords

isotonic regressionliquidity preference hypothesisorthantpolyhedral conepool adjacent violators algorithmpower

Mathematics Subject Classification ID

Parametric hypothesis testing (62F03) General nonlinear regression (62J02)

Abstract: We present a new way of testing ordered hypotheses against all alternatives which overpowers the classical approach both in simplicity and statistical power. Our new method tests the constrained likelihood ratio statistic against the quantile of one and only one chi-squared random variable with a data-dependent degrees of freedom instead of a mixture of chi-squares. Our new test is proved to have a valid finite-sample significance level $alpha$ and provides more power especially for sparse alternatives (those with a few or moderate number of null constraints violations) in comparison to the classical approach. Our method is also easier to use than the classical approach which requires to calculate or simulate a set of complicated weights. Two special cases are considered with more details, namely the case of testing orthants $m{u}_1<0,cdots,m{u}_n<0$ and the isotonic case of testing $m{u}_1<m{u}_2<m{u}_3$ against all alternatives. Contours of the difference in power are shown for these examples showing the interest of our new approach.

Recommendations

• Likelihood ratio tests with boundary constraints using data-dependent degrees of freedom
• scientific article; zbMATH DE number 3994740
• Likelihood ratio test for and against nonlinear inequality constraints
• Likelihood ratio test against a set of inequality constraints
• Admissibility of the Likelihood Ratio Test when the Parameter Space is Restricted under the Alternative

Cited in

(5)

• Simultaneous confidence intervals for ranks using the partitioning principle
• Empirical framework for two-player repeated games with random states
• The adaptive calibration of testingp-values
• Likelihood ratio tests with boundary constraints using data-dependent degrees of freedom
• Inference for the proportional odds cumulative logit model with monotonicity constraints for ordinal predictors and ordinal response

This page was built for publication: Adaptive critical value for constrained likelihood ratio testing

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5127213)