On testing marginal versus conditional independence

DOI10.1093/BIOMET/ASAA040zbMATH Open1457.62191arXiv1906.01850OpenAlexW2948870371MaRDI QIDQ5145695FDOQ5145695

Thomas S. Richardson, F. Richard Guo

Publication date: 21 January 2021

Published in: Biometrika (Search for Journal in Brave)

Abstract: We consider testing marginal independence versus conditional independence in a trivariate Gaussian setting. The two models are non-nested and their intersection is a union of two marginal independences. We consider two sequences of such models, one from each type of independence, that are closest to each other in the Kullback-Leibler sense as they approach the intersection. They become indistinguishable if the signal strength, as measured by the product of two correlation parameters, decreases faster than the standard parametric rate. Under local alternatives at such rate, we show that the asymptotic distribution of the likelihood ratio depends on where and how the local alternatives approach the intersection. To deal with this non-uniformity, we study a class of "envelope" distributions by taking pointwise suprema over asymptotic cumulative distribution functions. We show that these envelope distributions are well-behaved and lead to model selection procedures with rate-free uniform error guarantees and near-optimal power. To control the error even when the two models are indistinguishable, rather than insist on a dichotomous choice, the proposed procedure will choose either or both models.

Full work available at URL: https://arxiv.org/abs/1906.01850

Recommendations

• Testing for conditional multiple marginal independence
• Testing conditional independence via empirical likelihood
• Testing for independence in arbitrary distributions
• Testing conditional independence of discrete distributions
• scientific article; zbMATH DE number 5233421
• Test for conditional independence with application to conditional screening
• On a likelihood ratio test for independence
• Nonparametric tests for conditional independence using conditional distributions

zbMATH Keywords

model selectionGaussian graphical modellikelihood ratio testconditional independenceconfidenceenvelopecollider

Mathematics Subject Classification ID

Hypothesis testing in multivariate analysis (62H15) Probabilistic graphical models (62H22)

Cited In (11)

• Testing conditional independence using maximal nonlinear conditional correlation
• Testing independence between discrete random variables
• Testing conditional independence with data missing at random
• Testing for positive expectation dependence
• Test for conditional independence with application to conditional screening
• Universal inference
• A note on marginal and conditional independence
• Testing independence with additional information
• Title not available (Why is that?)
• Testing Conditional Independence Restrictions
• Testing conditional independence in supervised learning algorithms

This page was built for publication: On testing marginal versus conditional independence

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5145695)