Goodness-of-fit tests for high-dimensional Gaussian linear models

DOI10.1214/08-AOS629zbMATH Open1183.62074arXiv0711.2119OpenAlexW2075010327MaRDI QIDQ2380086FDOQ2380086

Authors: Nicolas Verzelen, Fanny Villers 

Publication date: 24 March 2010

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: Let $(Y,(X_i{)}_{iinmathcalI})$ be a zero mean Gaussian vector and $V$ be a subset of $mathcalI$. Suppose we are given $n$ i.i.d. replications of the vector $(Y,X)$. We propose a new test for testing that $Y$ is independent of $$ conditionally to $(X_i{)}_{iinV}$ against the general alternative that it is not. This procedure does not depend on any prior information on the covariance of $X$ or the variance of $Y$ and applies in a high-dimensional setting. It straightforwardly extends to test the neighbourhood of a Gaussian graphical model. The procedure is based on a model of Gaussian regression with random Gaussian covariates. We give non asymptotic properties of the test and we prove that it is rate optimal (up to a possible $log(n)$ factor) over various classes of alternatives under some additional assumptions. Besides, it allows us to derive non asymptotic minimax rates of testing in this setting. Finally, we carry out a simulation study in order to evaluate the performance of our procedure.

Full work available at URL: https://arxiv.org/abs/0711.2119

Recommendations

• Tests for Gaussian graphical models
• Adaptive tests of linear hypotheses by model selection
• High-dimensional Gaussian model selection on a Gaussian design
• A new test of linear hypothesis in regression
• A global homogeneity test for high-dimensional linear regression

zbMATH Keywords

Gaussian graphical modelslinear regressionmultiple testinggoodness-of-fitadaptive testingminimax hypothesis testingellipsoidminimax separation rate

Mathematics Subject Classification ID

Nonparametric hypothesis testing (62G10) Linear regression; mixed models (62J05) Applications of graph theory (05C90) Hypothesis testing in multivariate analysis (62H15) Measures of association (correlation, canonical correlation, etc.) (62H20)

Cites Work

• Gaussian Markov Random Fields
• Variable selection in high-dimensional linear models: partially faithful distributions and the PC-simple algorithm
• Least angle regression. (With discussion)
• The Dantzig selector: statistical estimation when \(p\) is much larger than \(n\). (With discussions and rejoinder).
• High-dimensional graphs and variable selection with the Lasso
• Title not available (Why is that?)
• Multiple testing and error control in Gaussian graphical model selection
• Regularization and Variable Selection Via the Elastic Net
• Tests for Gaussian graphical models
• Adaptive hypothesis testing using wavelets
• Model selection and estimation in the Gaussian graphical model
• Title not available (Why is that?)
• Title not available (Why is that?)
• The sparsity and bias of the LASSO selection in high-dimensional linear regression
• Adaptive estimation of a quadratic functional by model selection.
• Non-asymptotic minimax rates of testing in signal detection
• Estimation of Gaussian graphs by model selection
• Probabilistic Networks and Expert Systems
• Covariance matrix selection and estimation via penalised normal likelihood
• Concentration inequalities and model selection. Ecole d'Eté de Probabilités de Saint-Flour XXXIII -- 2003.
• Asymptotically minimax hypothesis testing for nonparametric alternatives. I
• Asymptotically minimax hypothesis testing for nonparametric alternatives. II
• Title not available (Why is that?)
• Asymptotically minimax hypothesis testing for nonparametric alternatives. III
• Adaptive tests of linear hypotheses by model selection
• Information-Theoretic Limits on Sparsity Recovery in the High-Dimensional and Noisy Setting
• Low-Order Conditional Independence Graphs for Inferring Genetic Networks

Cited In (23)

• Goodness-of-fit testing-based selection for large-\(p\)-small-\(n\) problems: a two-stage ranking approach
• Adaptive estimation of high-dimensional signal-to-noise ratios
• Detection boundary in sparse regression
• NN goodness-of-fit tests for linear models
• Multiple hypothesis testing for variable selection
• Combinatorial inference for graphical models
• A global homogeneity test for high-dimensional linear regression
• High-dimensional asymptotics of likelihood ratio tests in the Gaussian sequence model under convex constraints
• Optimal sparsity testing in linear regression model
• Detecting positive correlations in a multivariate sample
• Optimal estimation of variance in nonparametric regression with random design
• A consistent goodness-of-fit test for huge dimensional and functional data
• Minimax risks for sparse regressions: ultra-high dimensional phenomenons
• Efficient algorithm for testing goodness-of-fit for classification of high dimensional data
• High-dimensional Gaussian model selection on a Gaussian design
• Goodness‐of‐fit tests for the multivariate Student‐t distribution based on i.i.d. data, and for GARCH observations
• Classical testing in functional linear models
• Minimax adaptive tests for the functional linear model
• Goodness-of-Fit Tests for High Dimensional Linear Models
• Tests for Gaussian graphical models
• A joint test for parametric specification and independence in nonlinear regression models
• F-type testing in functional linear models
• Adaptive tests of linear hypotheses by model selection

Uses Software

• GMRFLib

This page was built for publication: Goodness-of-fit tests for high-dimensional Gaussian linear models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2380086)