Recent developments in high-dimensional inference for multivariate data: Parametric, semiparametric and nonparametric approaches

Solomon W. Harrar, Xiaoli Kong

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

In this paper, we give the most current account of methods for comparison of populations or treatment groups with high-dimensional data. We conveniently group the methods into three categories based on the hypothesis of interest and the model assumptions they make. We offer some perspectives on the connections and distinctions among the tests and discuss the ramifications of the model assumptions for practical applications. Among other things, we discuss the interpretation of the hypotheses and results of the appropriate tests and how this distinguishes the methods in terms of what data type they are suitable for. Further, we provide a discussion of computational complexity and a list of available R-packages implementations and their limitations. Finally, we illustrate the numerical performances of the various tests in a simulation study.

Original languageEnglish
Article number104855
JournalJournal of Multivariate Analysis
Volume188
DOIs
StatePublished - Mar 2022

Bibliographical note

Funding Information:
The authors are thankful to the Editor and Executive Editor for the efficient handling of the manuscript and their useful comments. The names of the authors are listed in alphabetic order.

Publisher Copyright:
© 2021 Elsevier Inc.

Keywords

  • High-dimensional data
  • Location test
  • Multivariate analysis
  • Nonparametric relative effect
  • Spatial rank
  • Spatial sign

ASJC Scopus subject areas

  • Statistics and Probability
  • Numerical Analysis
  • Statistics, Probability and Uncertainty

Fingerprint

Dive into the research topics of 'Recent developments in high-dimensional inference for multivariate data: Parametric, semiparametric and nonparametric approaches'. Together they form a unique fingerprint.

Cite this