Tail approximations for the Student t-, F-, and Welch statistics for non-normal and not necessarily i.i.d. random variables

DOI10.3150/13-BEJ552zbMATH Open1333.62053arXiv1410.6033OpenAlexW2950416934MaRDI QIDQ470067FDOQ470067

Authors: Dmitrii Zholud 

Publication date: 11 November 2014

Published in: Bernoulli (Search for Journal in Brave)

Abstract: Let $T$ be the Student one- or two-sample $t$-, $F$-, or Welch statistic. Now release the underlying assumptions of normality, independence and identical distribution and consider a more general case where one only assumes that the vector of data has a continuous joint density. We determine asymptotic expressions for $mathbfP(T>u)$ as $uoinfty$ for this case. The approximations are particularly accurate for small sample sizes and may be used, for example, in the analysis of High-Throughput Screening experiments, where the number of replicates can be as low as two to five and often extreme significance levels are used. We give numerous examples and complement our results by an investigation of the convergence speed - both theoretically, by deriving exact bounds for absolute and relative errors, and by means of a simulation study.

Full work available at URL: https://arxiv.org/abs/1410.6033

Recommendations

• scientific article; zbMATH DE number 5503197
• scientific article; zbMATH DE number 740442
• scientific article; zbMATH DE number 4078491
• Moderate deviations for two sample t-statistics
• Bounds for a nonparametric t table

zbMATH Keywords

outliersdependent random variablessmall sample sizetest power\(F\)-testhigh-throughput screeningnon-homogeneous datanon-normal population distributionStudent's one- and two-sample \(t\)-statisticssystematic effectsWelch statistics

Mathematics Subject Classification ID

Asymptotic distribution theory in statistics (62E20) Approximations to statistical distributions (nonasymptotic) (62E17)

Cites Work

• A Direct Approach to False Discovery Rates
• Edgeworth expansion for Student's t statistic under minimal moment conditions
• Strong Control, Conservative Point Estimation and Simultaneous Conservative Consistency of False Discovery Rates: A Unified Approach
• The positive false discovery rate: A Bayesian interpretation and the \(q\)-value
• Title not available (Why is that?)
• Title not available (Why is that?)
• Saddlepoint approximation for Student's \(t\)-statistic with no moment conditions
• Tail probability approximations for Student's \(t\)-statistics
• Title not available (Why is that?)
• Efficient estimation of the number of false positives in high-throughput screening
• Title not available (Why is that?)
• Tail approximations for the Student \(t\)-, \(F\)-, and Welch statistics for non-normal and not necessarily i.i.d. random variables
• THE DISTRIBUTION OF ‘STUDENT'S’ t IN RANDOM SAMPLES OF ANY SIZE DRAWN FROM NON-NORMAL UNIVERSES
• THE DISTRIBUTION OF THE VARIANCE RATIO IN RANDOM SAMPLES OF ANY SIZE DRAWN FROM NON-NORMAL UNIVERSES
• Corrections for Nonnormality in the Use of the Two-Sample $t$- and $F$-Test at High Significance Levels

Cited In (3)

• Title not available (Why is that?)
• Tail approximations for the Student \(t\)-, \(F\)-, and Welch statistics for non-normal and not necessarily i.i.d. random variables
• On the heavy-tailedness of Student's \(t\)-statistic

Uses Software

• Mathematica
• Matlab
• SmartTail

This page was built for publication: Tail approximations for the Student \(t\)-, \(F\)-, and Welch statistics for non-normal and not necessarily i.i.d. random variables

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q470067)