On the singular value distribution of large-dimensional data matrices whose columns have different correlations. Issue 2 (3rd March 2020)

• Record Type: Journal Article
• Authors: Yin, Yanqing
• Is Part Of: Statistics. Volume 54:Issue 2(2020)
• Page Start: 353
• Page End: 374
• Publication Date: 2020-03-03
• Subjects: Sample covariance matrix -- data matrix -- high-dimensional statistics -- random matrix theory -- LSD; Primary: 15B52 -- 60F15 -- 62E20 -- Secondary:60F17; Mathematical statistics -- Periodicals; 519.505
• DOI: 10.1080/02331888.2020.1720021
• Languages: English
• ISSNs: 0233-1888

Abstract

ABSTRACT: Suppose Y n = ( y 1, …, y n ) is a p × n data matrix whose columns y j, 1 ≤ j ≤ n have different correlations. The asymptotic spectral property of S n = ( 1 / n ) Y n Y n ∗ when p increases with n has recently been considered by some authors. This model has become increasingly popular because of its wide applications in multi-user multiple-input single-output (MISO) systems and robust signal processing. In this paper, for more convenient applications in practice, we will investigate the spectral distribution of S n under milder moment conditions than the existing work. Some applications of this model are also discussed.