Change search
ReferencesLink to record
Permanent link

Direct link
Spectral analysis of the Moore-Penrose inverse of a large dimensional sample covariance matrix
Stockholm University, Faculty of Science, Department of Mathematics.
2016 (English)In: Journal of Multivariate Analysis, ISSN 0047-259X, E-ISSN 1095-7243, Vol. 148, 160-172 p.Article in journal (Refereed) Published
Abstract [en]

For a sample of $n$ independent identically distributed $p$-dimensional centered random vectorswith covariance matrix $\bSigma_n$ let $\tilde{\bS}_n$ denote the usual sample covariance(centered by the mean) and $\bS_n$ the non-centered sample covariance matrix (i.e. the matrix of second moment estimates), where $p> n$. In this paper, we provide the limiting spectral distribution andcentral limit theorem for linear spectralstatistics of the Moore-Penrose inverse of $\bS_n$ and $\tilde{\bS}_n$. We consider the large dimensional asymptotics when the number of variables $p\rightarrow\infty$ and the sample size $n\rightarrow\infty$ such that $p/n\rightarrow c\in (1, +\infty)$. We present a Marchenko-Pastur law for both types of matrices, which shows that the limiting spectral distributions for both sample covariance matrices are the same. On the other hand, we demonstrate that the asymptotic distribution of linear spectral statistics of the Moore-Penrose inverse of $\tilde{\bS}_n$ differs in the mean from that of $\bS_n$.

Place, publisher, year, edition, pages
2016. Vol. 148, 160-172 p.
Keyword [en]
CLT, large-dimensional asymptotics, Moore-Penrose inverse, random matrix theory
National Category
Probability Theory and Statistics
Research subject
URN: urn:nbn:se:su:diva-130903DOI: 10.1016/j.jmva.2016.03.001ISI: 000375826400012OAI: diva2:934088
Available from: 2016-06-08 Created: 2016-06-08 Last updated: 2016-06-14Bibliographically approved

Open Access in DiVA

No full text

Other links

Publisher's full text

Search in DiVA

By author/editor
Bodnar, Taras
By organisation
Department of Mathematics
In the same journal
Journal of Multivariate Analysis
Probability Theory and Statistics

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Altmetric score

Total: 10 hits
ReferencesLink to record
Permanent link

Direct link