Univariate likelihood projections and characterizations of the multivariate normal distribution

DOI10.1016/J.JMVA.2020.104643zbMATH Open1448.62028arXiv1910.12394OpenAlexW3035340491MaRDI QIDQ2196125FDOQ2196125

Authors: Albert Vexler 

Publication date: 28 August 2020

Published in: Journal of Multivariate Analysis (Search for Journal in Brave)

Abstract: The problem of characterizing a multivariate distribution of a random vector using examination of univariate combinations of vector components is an essential issue of multivariate analysis. The likelihood principle plays a prominent role in developing powerful statistical inference tools. In this context, we raise the question: can the univariate likelihood function based on a random vector be used to provide the uniqueness in reconstructing the vector distribution? In multivariate normal (MN) frameworks, this question links to a reverse of Cochran's theorem that concerns the distribution of quadratic forms in normal variables. We characterize the MN distribution through the univariate likelihood type projections. The proposed principle is employed to illustrate simple techniques for assessing multivariate normality via well-known tests that use univariate observations. The presented testing strategy can exhibit high and stable power characteristics in comparison to the well-known procedures in various scenarios when observed vectors are non-MN distributed, whereas their components are normally distributed random variables. In such cases, the classical multivariate normality tests may break down completely. KEY WORDS: Characterization, Goodness of fit, Infinity divisible, Likelihood, Multivariate normal distribution, Projection, Quadratic form, Test for multivariate normality.

Full work available at URL: https://arxiv.org/abs/1910.12394

Recommendations

• Likelihood ratio tests for multivariate normality
• A characterization of multivariate normal distribution and its application
• A test of multivariate normality based on likelihood functions
• Testing multinormality based on low-dimensional projection
• Testing for normality in arbitrary dimension

zbMATH Keywords

likelihoodmultivariate normal distributionprojectionquadratic formtest for multivariate normalitycharacterizationinfinity divisible

Mathematics Subject Classification ID

Probability distributions: general theory (60E05) Infinitely divisible distributions; stable distributions (60E07) Nonparametric hypothesis testing (62G10) Multivariate distribution of statistics (62H10) Hypothesis testing in multivariate analysis (62H15) Characterization and structure theory of statistical distributions (62E10)

Cites Work

• Smooth Tests of Goodness of Fit
• Towards data driven selection of a penalty function for data driven Neyman tests
• Data-Driven Version of Neyman's Smooth Test of Fit
• Invariant tests for multivariate normality: A critical review
• A consistent test for multivariate normality based on the empirical characteristic function
• A test for multivariate normality based on sample entropy and projection pursuit
• The non-singularity of generalized sample covariance matrices
• Title not available (Why is that?)
• Title not available (Why is that?)
• Likelihood ratio tests for multivariate normality
• Data-Driven Rank Tests for Independence
• A characterization of multivariate normality through univariate projections
• Springer handbook of engineering statistics. With CD-ROM.
• The Distribution of a Quadratic Form of Normal Random Variables
• A contemporary review and bibliography of infinitely divisible distributions and processes
• A characterization of the Gaussian distribution
• A characterisation of the Gaussian distribution through the sample variance
• Infinitely divisible distributions, conditions for independence, and central limit theorems
• Notes on the distribution of quadratic forms in singular normal variables
• On multivariate infinitely divisible distributions
• Infinitely divisible measures with independent marginals
• Data driven smooth tests for bivariate normality
• Random and vector measures.
• Statistics in the health sciences. Theory, applications, and computing
• Title not available (Why is that?)
• Title not available (Why is that?)
• On the Problem of Reconstructing a Summands Distribution by the Distribution of Their Sum
• Letter to the editor

Cited In (2)

• A comment on affine invariance and ancillarity in testing multivariate normality
• An exact projection pursuit-based algorithm for multivariate two-sample nonparametric testing applicable to retrospective and group sequential studies

Uses Software

• R
• testforDEP

This page was built for publication: Univariate likelihood projections and characterizations of the multivariate normal distribution

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2196125)