Change Detection in The Covariance Structure of High-Dimensional Gaussian Low-Rank Models
Langue
EN
Communication dans un congrès
Ce document a été publié dans
2021 IEEE Statistical Signal Processing Workshop (SSP), 2021-07-11, Rio de Janeiro. 2021-07-11
IEEE
Résumé en anglais
This paper is devoted to the problem of testing equality between the covariance matrices of L multivariate Gaussian time series with dimension M, in the context where each of the L covariance matrices is the sum of a ...Lire la suite >
This paper is devoted to the problem of testing equality between the covariance matrices of L multivariate Gaussian time series with dimension M, in the context where each of the L covariance matrices is the sum of a low-rank K component and the identity matrix. Assuming N 1 , …, N L samples are available for each time series, a new test statistic, based on the eigenvalues of the L sample covariance matrices (SCM) of each time series as well as the eigenvalues of a pooled SCM mixing the N 1 +…+N L available samples, is pro-posed and proved to be consistent in the high dimensional regime in which M, N 1 , …, N L converge to infinity at the same rate, while K and L are kept fixed. Numerical simulations show that the proposed test statistic is competitive with other relevant methods for moderate values of M, N 1 , …, N L .< Réduire
Mots clés en anglais
Change detection
Covariance
Spiked models
Random matrix theory
Unités de recherche