Asymptotic power of sphericity tests for high-dimensional data

DOI10.1214/13-AOS1100zbMATH Open1293.62125arXiv1306.4867MaRDI QIDQ366967FDOQ366967

Authors: Alexei Onatski, Marcelo J. Moreira, Marc Hallin 

Publication date: 25 September 2013

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

Abstract: This paper studies the asymptotic power of tests of sphericity against perturbations in a single unknown direction as both the dimensionality of the data and the number of observations go to infinity. We establish the convergence, under the null hypothesis and contiguous alternatives, of the log ratio of the joint densities of the sample covariance eigenvalues to a Gaussian process indexed by the norm of the perturbation. When the perturbation norm is larger than the phase transition threshold studied in Baik, Ben Arous and Peche [Ann. Probab. 33 (2005) 1643-1697] the limiting process is degenerate, and discrimination between the null and the alternative is asymptotically certain. When the norm is below the threshold, the limiting process is nondegenerate, and the joint eigenvalue densities under the null and alternative hypotheses are mutually contiguous. Using the asymptotic theory of statistical experiments, we obtain asymptotic power envelopes and derive the asymptotic power for various sphericity tests in the contiguity region. In particular, we show that the asymptotic power of the Tracy-Widom-type tests is trivial (i.e., equals the asymptotic size), whereas that of the eigenvalue-based likelihood ratio test is strictly larger than the size, and close to the power envelope.

Full work available at URL: https://arxiv.org/abs/1306.4867

Recommendations

• Some sphericity tests for high dimensional data based on ratio of the traces of sample covariance matrices
• On the sphericity test with large-dimensional observations
• Sphericity test for high dimensional data based on random matrix theory
• High-dimensional sphericity test by extended likelihood ratio
• A new test for sphericity of the covariance matrix for high dimensional data

zbMATH Keywords

asymptotic powercontiguitypower envelopesteepest descentcontour integral representationlarge dimensionalitysphericity testsspiked covariance

Mathematics Subject Classification ID

Hypothesis testing in multivariate analysis (62H15) Theory of statistical experiments (62B15) Asymptotic approximations, asymptotic expansions (steepest descent, etc.) (41A60)

Cites Work

• Asymptotic Statistics
• Laplace approximations for hypergeometric functions with matrix argument
• On the distribution of the largest eigenvalue in principal components analysis
• Semiparametrically efficient rank-based inference for shape. I: optimal rank-based tests for sphericity
• Eigenvalues of large sample covariance matrices of spiked population models
• A note on testing the covariance matrix for large dimension
• Title not available (Why is that?)
• The distribution of a statistic used for testing sphericity of normal distributions
• Finite sample approximation results for principal component analysis: A matrix perturbation approach
• Sample Eigenvalue Based Detection of High-Dimensional Signals in White Noise Using Relatively Few Samples
• Title not available (Why is that?)
• Testing hypotheses about the number of factors in large factor models
• A new test for sphericity of the covariance matrix for high dimensional data
• Some hypothesis tests for the covariance matrix when the dimension is large compared to the sample size
• A high-dimensional test for the equality of the smallest eigenvalues of a covariance matrix
• Optimal detection of sparse principal components in high dimension
• Tests for high-dimensional covariance matrices
• Convergence rate of expected spectral distributions of large random matrices. II: Sample covariance matrices
• Corrections to LRT on large-dimensional covariance matrix by RMT
• Tracy-Widom limit for the largest eigenvalue of a large class of complex sample covariance matrices
• Phase transition of the largest eigenvalue for nonnull complex sample covariance matrices
• Limits of spiked random matrices. I
• Means of a Dirichlet process and multiple hypergeometric functions.
• CLT for linear spectral statistics of large-dimensional sample covariance matrices.
• Multiple Hypergeometric Functions: Probabilistic Interpretations and Statistical Uses
• On the empirical distribution of eigenvalues of a class of large dimensional random matrices
• Title not available (Why is that?)
• Asymptotic power of sphericity tests for high-dimensional data
• A Fourier view on the \(R\)-transform and related asymptotics of spherical integrals
• Distributions of Matrix Variates and Latent Roots Derived from Normal Samples
• The largest eigenvalues of sample covariance matrices for a spiked population: diagonal case
• Rank 1 real Wishart spiked model
• Title not available (Why is that?)
• On singular Wishart and singular multivariate beta distributions
• The density of a quadratic form in a vector uniformly distributed on the \(n\)-sphere
• Automatic PCA dimension selection for high dimensional data and small sample sizes
• The largest eigenvalue of real symmetric, Hermitian and Hermitian self-dual random matrix models with rank one external source. I
• Non-Parametric Detection of the Number of Signals: Hypothesis Testing and Random Matrix Theory
• Performance of Statistical Tests for Single-Source Detection Using Random Matrix Theory
• Locally Best Invariant Test for Sphericity and the Limiting Distributions
• Title not available (Why is that?)
• Significance Test for Sphericity of a Normal $n$-Variate Distribution

Cited In (65)

• Contiguity under high-dimensional Gaussianity with applications to covariance testing
• Subexponential-time algorithms for sparse PCA
• Distance correlation test for high-dimensional independence
• Hypergeometric functions of matrix arguments and linear statistics of multi-spiked Hermitian matrix models
• Large deviation principles induced by the Stiefel manifold, and random multidimensional projections
• Long random matrices and tensor unfolding
• A CLT for the LSS of large-dimensional sample covariance matrices with diverging spikes
• An edge CLT for the log determinant of Wigner ensembles
• A note on the CLT of the LSS for sample covariance matrix from a spiked population model
• Statistical limits of spiked tensor models
• Fundamental limits of detection in the spiked Wigner model
• High-dimensional covariance matrices in elliptical distributions with application to spherical test
• Phase transition in the spiked random tensor with Rademacher prior
• Fluctuations of the overlap at low temperature in the 2-spin spherical SK model
• On high-dimensional sign tests
• Optimal signal detection in some spiked random matrix models: likelihood ratio tests and linear spectral statistics
• Rapid evaluation of the spectral signal detection threshold and Stieltjes transform
• Statistical and computational limits for sparse matrix detection
• Testing in high-dimensional spiked models
• Asymptotic power of sphericity tests for high-dimensional data
• Likelihood ratio tests for high-dimensional normal distributions
• Central limit theorems for classical likelihood ratio tests for high-dimensional normal distributions
• Optimal hypothesis testing for high dimensional covariance matrices
• A necessary and sufficient condition for edge universality at the largest singular values of covariance matrices
• Asymptotic non-null distribution for the locally most powerful invariant test for sphericity
• Some new results on the eigenvalues of complex non-central Wishart matrices with a rank-1 mean
• The spectral norm of random inner-product kernel matrices
• Estimating structured high-dimensional covariance and precision matrices: optimal rates and adaptive estimation
• Detection of weak signals in high-dimensional complex-valued data
• Testing the eigenvalue structure of spot and integrated covariance
• Asymptotic power of the sphericity test under weak and strong factors in a fixed effects panel data model
• Power computation for hypothesis testing with high-dimensional covariance matrices
• A generalized likelihood ratio test for normal mean when \(p\) is greater than \(n\)
• On the sphericity test with large-dimensional observations
• Statistical inference for principal components of spiked covariance matrices
• Estimation of functionals of sparse covariance matrices
• Tests for covariance matrices in high dimension with less sample size
• Phase transition in random tensors with multiple independent spikes
• Edge universality of separable covariance matrices
• A robust test for sphericity of high-dimensional covariance matrices
• Projection tests for high-dimensional spiked covariance matrices
• High-dimensional linear models: a random matrix perspective
• Signal detection in high dimension: the multispiked case
• Testing uniformity on high-dimensional spheres: the non-null behaviour of the Bingham test
• Random matrix theory and its applications
• Some sphericity tests for high dimensional data based on ratio of the traces of sample covariance matrices
• Pythagorean generalization of testing the equality of two symmetric positive definite matrices
• Optimal prediction in the linearly transformed spiked model
• Permutation methods for factor analysis and PCA
• Asymptotic power of likelihood ratio tests for high dimensional data
• Comparison of powers for the sphericity tests using both the asymptotic distribution and the bootstrap
• Optimal two-sided tests for instrumental variables regression with heteroskedastic and autocorrelated errors
• Hypothesis testing for high-dimensional covariance matrices
• Asymptotics of the principal components estimator of large factor models with weakly influential factors
• Universality for the largest eigenvalue of sample covariance matrices with general population
• Asymptotic analysis of the squared estimation error in misspecified factor models
• Free energy fluctuations of the two-spin spherical SK model at critical temperature
• Testing for sphericity in a fixed effects panel data model with time-varying variances
• High-dimensional rank tests for sphericity
• Power in high-dimensional testing problems
• Asymptotic joint distribution of extreme eigenvalues and trace of large sample covariance matrix in a generalized spiked population model
• Optimal detection of sparse principal components in high dimension
• High-dimensional tests for spherical location and spiked covariance
• Optimality and sub-optimality of PCA. I: Spiked random matrix models
• Asymptotic linear spectral statistics for spiked Hermitian random matrices

This page was built for publication: Asymptotic power of sphericity tests for high-dimensional data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q366967)