☆ 4.6 Article

Two-sample test of high dimensional means under dependence

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY (2014)

期刊

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY

卷 76, 期 2, 页码 349-372

出版社

WILEY-BLACKWELL

DOI: 10.1111/rssb.12034

关键词

Covariance matrix; Extreme value distribution; High dimensional test; Hypothesis testing; Limiting null distribution; Power; Precision matrix; Testing equality of mean vectors

类别

Statistics & Probability

资金

National Science Foundation Focused Research Group [DMS-0854973]
National Natural Science Foundation of China [11201298]
Program for Professor of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning
'Foundation for the author of national excellent doctoral dissertation of PR China'
Shanghai Jiao Tong University, People's Republic of China
Division Of Mathematical Sciences
Direct For Mathematical & Physical Scien [1208982] Funding Source: National Science Foundation

向作者/读者索取更多资源

Protocol

Reagent

摘要

The paper considers in the high dimensional setting a canonical testing problem in multivariate analysis, namely testing the equality of two mean vectors. We introduce a new test statistic that is based on a linear transformation of the data by the precision matrix which incorporates the correlations between the variables. The limiting null distribution of the test statistic and the power of the test are analysed. It is shown that the test is particularly powerful against sparse alternatives and enjoys certain optimality. A simulation study is carried out to examine the numerical performance of the test and to compare it with other tests given in the literature. The results show that the test proposed significantly outperforms those tests in a range of settings.

作者

我是这篇论文的作者

点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.6

评分不足

次要评分

新颖性

-

重要性

-

科学严谨性

-

评价这篇论文

推荐

Article Statistics & Probability

Fast nonasymptotic testing and support recovery for large sparse Toeplitz covariance matrices

Nayel Bettache, Cristina Butucea, Marianne Sorba

Summary: This study examines n independent p-dimensional Gaussian vectors with covariance matrix having a Toeplitz structure, aiming to test the independence of the vectors and select the support of non-zero entries. Test procedures are developed, showing non-asymptotic behavior under one-sided and two-sided alternatives. Results also extend to nearly Toeplitz covariance structure and sub-Gaussian vectors, with numerical results indicating excellent performance of the test procedures and support selectors, particularly as dimension p increases.

JOURNAL OF MULTIVARIATE ANALYSIS (2022)

添加到收藏夹

Article Biology

Nonparametric cluster significance testing with reference to a unimodal null distribution

Erika S. Helgeson, David M. Vock, Eric Bair

Summary: This paper proposes a novel method to evaluate the significance of identified clusters by comparing the explained variation due to clustering from the original data to a unimodal reference distribution that preserves the covariance structure in the data. The approach is adapted for high-dimension low-sample size settings and can be used to test the null hypothesis and determine the optimal number of clusters.

BIOMETRICS (2021)

添加到收藏夹

Editorial Material Obstetrics & Gynecology

Current controversies: Null hypothesis significance testing

Philip M. Sedgwick, Anne Hammer, Ulrik Schioler Kesmodel, Lars Henning Pedersen

Summary: Traditional null hypothesis significance testing (NHST) is widely used in obstetric and gynecological research, but its application in inferring clinical significance is controversial. Misinterpretation of statistical significance and ignorance of NHST limitations may lead to false claims and dismissal of important factors.

ACTA OBSTETRICIA ET GYNECOLOGICA SCANDINAVICA (2022)

添加到收藏夹

Article Psychology, Multidisciplinary

Prior Sensitivity of Null Hypothesis Bayesian Testing

Herbert Hoijtink

Summary: Researchers increasingly use Bayes factor for hypotheses evaluation, with NHBT being sensitive to the specification of the scale parameter of the prior distribution, while IHBT is not. Using recommended default values for scaling parameters in NHBT leads to unpredictable operation characteristics, but selecting the scaling parameter to bias the Bayes factor towards the null hypothesis by 19 if the observed effect size is zero can address this issue in some cases. However, this does not solve all problems associated with NHBT.

PSYCHOLOGICAL METHODS (2022)

添加到收藏夹

Article Economics

Testing high-dimensional covariance matrices under the elliptical distribution and beyond

Xinxin Yang, Xinghua Zheng, Jiaqi Chen

Summary: Tests for high-dimensional covariance matrices based on a generalized elliptical model are developed, without assuming specific parametric distributions or involving data kurtosis. These tests can be used to test uncorrelatedness among idiosyncratic returns, demonstrating their flexibility and applicability.

JOURNAL OF ECONOMETRICS (2021)

添加到收藏夹

Article Statistics & Probability

Covariance matrix testing in high dimension using random projections

Deepak Nag Ayyala, Santu Ghosh, Daniel F. Linder

Summary: This paper proposes a test procedure called CRAMP for testing hypotheses involving one or more covariance matrices using random projections, which alleviates the curse of dimensionality in high dimensions. Through randomly projecting high-dimensional data into lower-dimensional subspaces, traditional multivariate tests can be used effectively.

COMPUTATIONAL STATISTICS (2022)

添加到收藏夹

Article Multidisciplinary Sciences

Beta distribution misspecification tests with application to Covid-19 mortality rates in the United States

Jose Jairo Santana-e-Silva, Francisco Cribari-Neto, Klaus L. P. Vasconcellos

Summary: The beta distribution is commonly used to model variables in (0, 1). This paper examines its adequacy and tests the hypothesis using information matrix equality. The results show that the beta distribution performs well in representing Covid-19 death rates, especially when considering the impact of vaccination.

PLOS ONE (2022)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

BFpack: Flexible Bayes Factor Testing of Scientific Theories in R

Joris Mulder, Donald R. Williams, Xin Gu, Andrew Tomarken, Florian Boing-Messing, Anton Olsson-Collentine, Marlyne Meijerink, Janosch Menke, Robbie van Aert, Jean-Paul Fox, Herbert Hoijtink, Yves Rosseel, Eric-Jan Wagenmakers, Caspar van Lissa

Summary: There have been significant methodological developments in Bayes factors for hypothesis testing in the social and behavioral sciences, but available software tools are still limited. BFpack is a new R package that offers Bayes factor hypothesis testing functions for common testing problems.

JOURNAL OF STATISTICAL SOFTWARE (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

Flexible Robust Adaptive Beamforming Method With Null Widening

Ziwei Liu, Shanshan Zhao, Chen Zhang, Gengxin Zhang

Summary: The nonstationarity of interferences and array errors can significantly degrade the interference cancellation performance of an adaptive beamformer. Covariance matrix tapering (CMT) can create wide troughs in the receiving pattern and is a promising solution. However, the need for symmetrical and equivalent-width nulls in adaptive patterns is often unnecessary, as different interferences are unlikely to have the same nonstationarity. This paper proposes a flexible method for null widening that can produce wide null with different desired width and asymmetry, which is based on spatial asymmetrical interference cluster and different tapering matrices.

IEEE SENSORS JOURNAL (2021)

添加到收藏夹

Article Statistics & Probability

HYPOTHESIS TESTING FOR THE COVARIANCE MATRIX IN HIGH-DIMENSIONAL TRANSPOSABLE DATA WITH KRONECKER PRODUCT DEPENDENCE STRUCTURE

Anestis Touloumis, John C. Marioni, Simon Tavare

Summary: The matrix-variate normal distribution is a popular model for high-dimensional transposable data. There is a proposed testing methodology for assessing the covariance matrices in high-dimensional settings, which shows robustness to normality departures and good statistical power against alternative hypotheses. The utility of the proposed tests is demonstrated through analysis of microarray and electroencephalogram studies.

STATISTICA SINICA (2021)

添加到收藏夹

Article Statistics & Probability

Testing for the sandwich-form covariance matrix of the quasi-maximum likelihood estimator

Lijuan Huo, Jin Seo Cho

Summary: This study tests the sandwich-form asymptotic covariance matrices under conditionally heteroskedastic and/or autocorrelated regression errors, providing a testing methodology to detect their influence. A sequential testing procedure is established to achieve the research goal, confirming the theory through simulation and empirical data.

TEST (2021)

添加到收藏夹

Article Mathematical & Computational Biology

Testing a global null hypothesis using ensemble machine learning methods

Sunwoo Han, Youyi Fong, Ying Huang

Summary: Testing the global null hypothesis that there are no significant predictors for a binary outcome of interest among a large set of biomarker measurements is crucial in biomedical studies. This paper proposes to enhance the power of such testing methods by utilizing ensemble machine learning techniques. The effectiveness of the proposed methods is demonstrated through Monte Carlo studies and the application to immunologic biomarkers dataset from the RV144 HIV vaccine efficacy trial.

STATISTICS IN MEDICINE (2022)

添加到收藏夹

Article Health Care Sciences & Services

A logical analysis of null hypothesis significance testing using popular terminology

Richard McNulty

Summary: NHST's internal logic can be analyzed using propositional calculus, with the testable H-0 determined by analyzing the range of P-values; The correspondence between H-0 and H-A must be exhaustive to avoid false dichotomies; The conclusions derived from NHST only justify that the results are not due to chance alone, rather than proving the research hypothesis is true.

BMC MEDICAL RESEARCH METHODOLOGY (2022)

添加到收藏夹

Article Mathematics, Applied

Limiting spectral distribution of high-dimensional noncentral Fisher matrices and its analysis

Xiaozhuo Zhang, Zhidong Bai, Jiang Hu

Summary: This paper investigates the limiting spectral distribution and analytic behavior of high-dimensional noncentral Fisher matrices and presents the determination criterion for the support of the limiting spectral distribution.

SCIENCE CHINA-MATHEMATICS (2023)

添加到收藏夹

Article Biochemical Research Methods

Identifying pleiotropic genes for complex phenotypes with summary statistics from a perspective of composite null hypothesis testing

Ting Wang, Haojie Lu, Ping Zeng

Summary: Pleiotropy is important for understanding the genetic connection between complex phenotypes and diseases. This study proposes a gene-based method called MAIUP for efficient pleiotropy identification, considering the high-dimensional composite null hypothesis. MAIUP takes into account the composite nature of pleiotropy test and provides well-calibrated P-values for controlling family-wise error rate and false discovery rate. It also effectively addresses the issue of overlapping subjects commonly encountered in association studies. Simulation studies demonstrate the superiority of MAIUP in maintaining correct type I error control and higher power compared to other methods. Application of MAIUP to psychiatric disorders discovers new pleiotropic genes and functional and enrichment analyses support their association with the disorders.

BRIEFINGS IN BIOINFORMATICS (2022)

添加到收藏夹

Article Statistics & Probability

ESTIMATING SPARSE PRECISION MATRIX: OPTIMAL RATES OF CONVERGENCE AND ADAPTIVE ESTIMATION

T. Tony Cai, Weidong Liu, Harrison H. Zhou

ANNALS OF STATISTICS (2016)

添加到收藏夹

Article Statistics & Probability

Large-Scale Multiple Testing of Correlations

T. Tony Cai, Weidong Liu

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION (2016)

添加到收藏夹

Article Statistics & Probability

JOINT ESTIMATION OF MULTIPLE HIGH-DIMENSIONAL PRECISION MATRICES

T. Tony Cai, Hongzhe Li, Weidong Liu, Jichun Xie

STATISTICA SINICA (2016)

添加到收藏夹

Article Rheumatology

Phenome-Wide Association Study of Autoantibodies to Citrullinated and Noncitrullinated Epitopes in Rheumatoid Arthritis

Katherine P. Liao, Jeffrey A. Sparks, Boris P. Hejblum, I-Hsin Kuo, Jing Cui, Lauren J. Lahey, Andrew Cagan, Vivian S. Gainer, Weidong Liu, T. Tony Cai, Jeremy Sokolove, Tianxi Cai

ARTHRITIS & RHEUMATOLOGY (2017)

添加到收藏夹

Article Statistics & Probability

TESTING INDEPENDENCE WITH HIGH-DIMENSIONAL CORRELATED SAMPLES

Xi Chen, Weidong Liu

ANNALS OF STATISTICS (2018)

添加到收藏夹

Article Statistics & Probability

Simultaneous nonparametric regression analysis of sparse longitudinal data

Hongyuan Cao, Weidong Liu, Zhou Zhou

BERNOULLI (2018)

添加到收藏夹

Article Statistics & Probability

SIMULTANEOUS CONFIDENCE BANDS IN NONLINEAR REGRESSION MODELS WITH NONSTATIONARITY

Degui Li, Weidong Liu, Qiying Wang, Wei Biao Wu

STATISTICA SINICA (2017)

添加到收藏夹

Article Statistics & Probability

GRAPH ESTIMATION FOR MATRIX-VARIATE GAUSSIAN DATA

Xi Chen, Weidong Liu

STATISTICA SINICA (2019)

添加到收藏夹

Article Statistics & Probability

QUANTILE REGRESSION UNDER MEMORY CONSTRAINT

Xi Chen, Weidong Liu, Yichen Zhang

ANNALS OF STATISTICS (2019)

添加到收藏夹

Article Statistics & Probability

First-Order Newton-Type Estimator for Distributed Estimation and Inference

Xi Chen, Weidong Liu, Yichen Zhang

Summary: This article investigates distributed estimation and inference for a general statistical problem by proposing a new multi-round distributed estimation procedure to overcome restrictions on the number of machines. The method introduces a computationally efficient estimator for Sigma(-1)w, applicable to nondifferentiable losses, facilitating inference for the empirical risk minimizer.

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Byzantine-robust distributed sparse learning for M-estimation

Jiyuan Tu, Weidong Liu, Xiaojun Mao

Summary: This paper presents a Byzantine-resilient method for distributed sparse M-estimation. By constructing a pseudo-response variable and transforming the optimization problem, a communication-efficient distributed algorithm is developed. Theoretically, it is proven that the proposed method achieves fast convergence and a support recovery result is established.

MACHINE LEARNING (2023)

添加到收藏夹

Article Automation & Control Systems

Distributed High-dimensional Regression Under a Quantile Loss Function

Xi Chen, Weidong Liu, Xiaojun Mao, Zhuoyi Yang

JOURNAL OF MACHINE LEARNING RESEARCH (2020)

添加到收藏夹

Article Statistics & Probability

Spectral analysis of high-dimensional time series

Mark Fiecas, Chenlei Leng, Weidong Liu, Yi Yu

ELECTRONIC JOURNAL OF STATISTICS (2019)

添加到收藏夹

Article Statistics & Probability

LARGE-SCALE SIMULTANEOUS TESTING OF CROSS-COVARIANCE MATRICES WITH APPLICATIONS TO PheWAS

Tianxi Cai, T. Tony Cai, Katherine Liao, Weidong Liu

STATISTICA SINICA (2019)

添加到收藏夹

Article Statistics & Probability

STRUCTURAL SIMILARITY AND DIFFERENCE TESTING ON MULTIPLE SPARSE GAUSSIAN GRAPHICAL MODELS

Weidong Liu

ANNALS OF STATISTICS (2017)

添加到收藏夹

暂无数据

© Peeref 2019-2024. All rights reserved.