arXiv:1708.06296 [math.PR]AbstractReferencesReviewsResources
Asymptotics of empirical eigen-structure for high dimensional sample covariance matrices of general form
Published 2017-08-21Version 1
In this paper, we study the local asymptotics of the eigenvalues and eigenvectors for a general class of sample covariance matrices, where the spectrum of the population covariance matrices can have a finite number of spikes and bulk components. Our paper is a unified framework combining the spiked model and covariance matrices without outliers. Examples and statistical applications are considered to illustrate our results.
Categories: math.PR
Related articles: Most relevant | Search more
arXiv:1612.06835 [math.PR] (Published 2016-12-20)
Box constrained $\ell_1$ optimization in random linear systems -- asymptotics
arXiv:1607.07636 [math.PR] (Published 2016-07-26)
Asymptotics for the Time of Ruin in the War of Attrition
arXiv:math/0308157 [math.PR] (Published 2003-08-16)
On Asymptotics for the Airy Process