Salvato in:
Dettagli Bibliografici
Autori principali: Schüürhuis, Stephen, Konietschke, Frank, Brunner, Edgar
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2504.01796
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916678176079872
author Schüürhuis, Stephen
Konietschke, Frank
Brunner, Edgar
author_facet Schüürhuis, Stephen
Konietschke, Frank
Brunner, Edgar
contents We propose a new test to address the nonparametric Behrens-Fisher problem involving different distribution functions in the two samples. Our procedure tests the null hypothesis $\mathcal{H}_0: θ= \frac{1}{2}$, where $θ= P(X<Y) + \frac{1}{2}P(X=Y)$ denotes the Mann-Whitney effect. No restrictions on the underlying distributions of the data are imposed with the trivial exception of one-point distributions. The method is based on evaluating the ratio of the variance $σ_N^2$ of the Mann-Whitney effect estimator $\widehatθ$ to its theoretical maximum, as derived from the Birnbaum-Klose inequality. Through simulations, we demonstrate that the proposed test effectively controls the type-I error rate under various conditions, including small sample sizes, unbalanced designs, and different data-generating mechanisms. Notably, it provides better control of the type-1 error rate compared to the widely used Brunner-Munzel test, particularly at small significance levels such as $α\in \{0.01, 0.005\}$. Additionally, we derive range-preserving compatible confidence intervals, showing that they offer improved coverage over those compatible to the Brunner-Munzel test. Finally, we illustrate the application of our method in a clinical trial example.
format Preprint
id arxiv_https___arxiv_org_abs_2504_01796
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A New Approach to the Nonparametric Behrens-Fisher Problem with Compatible Confidence Intervals
Schüürhuis, Stephen
Konietschke, Frank
Brunner, Edgar
Methodology
We propose a new test to address the nonparametric Behrens-Fisher problem involving different distribution functions in the two samples. Our procedure tests the null hypothesis $\mathcal{H}_0: θ= \frac{1}{2}$, where $θ= P(X<Y) + \frac{1}{2}P(X=Y)$ denotes the Mann-Whitney effect. No restrictions on the underlying distributions of the data are imposed with the trivial exception of one-point distributions. The method is based on evaluating the ratio of the variance $σ_N^2$ of the Mann-Whitney effect estimator $\widehatθ$ to its theoretical maximum, as derived from the Birnbaum-Klose inequality. Through simulations, we demonstrate that the proposed test effectively controls the type-I error rate under various conditions, including small sample sizes, unbalanced designs, and different data-generating mechanisms. Notably, it provides better control of the type-1 error rate compared to the widely used Brunner-Munzel test, particularly at small significance levels such as $α\in \{0.01, 0.005\}$. Additionally, we derive range-preserving compatible confidence intervals, showing that they offer improved coverage over those compatible to the Brunner-Munzel test. Finally, we illustrate the application of our method in a clinical trial example.
title A New Approach to the Nonparametric Behrens-Fisher Problem with Compatible Confidence Intervals
topic Methodology
url https://arxiv.org/abs/2504.01796