Spectral analysis of the Moore-Penrose inverse of a large dimensional sample covariance matrix
نویسندگان
چکیده
For a sample of n independent identically distributed p-dimensional centered random vectors with covariance matrix Σn let S̃n denote the usual sample covariance (centered by the mean) and Sn 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 and central limit theorem for linear spectral statistics of the Moore-Penrose inverse of Sn and S̃n. We consider the large dimensional asymptotics when the number of variables p → ∞ and the sample size n → ∞ such that p/n → c ∈ (1,+∞). 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 S̃n differs in the mean from that of Sn. AMS 2010 subject classifications: 60B20, 60F05, 60F15, 60F17, 62H10
منابع مشابه
An Efficient Schulz-type Method to Compute the Moore-Penrose Inverse
A new Schulz-type method to compute the Moore-Penrose inverse of a matrix is proposed. Every iteration of the method involves four matrix multiplications. It is proved that this method converge with fourth-order. A wide set of numerical comparisons shows that the average number of matrix multiplications and the average CPU time of our method are considerably less than those of other methods.
متن کاملComparison of Discrimination Methods for High Dimensional Data
In microarray experiments, the dimension p of the data is very large but there are only few observations N on the subjects/patients. In this article, the problem of classifying a subject into one of the two groups, when p is large, is considered. Three procedures based on Moore-Penrose inverse of the sample covariance matrix and an empirical Bayes estimate of the precision matrix are proposed a...
متن کاملHermitian solutions to the system of operator equations T_iX=U_i.
In this article we consider the system of operator equations T_iX=U_i for i=1,2,...,n and give necessary and suffcient conditions for the existence of common Hermitian solutions to this system of operator equations for arbitrary operators without the closedness condition. Also we study the Moore-penrose inverse of a ncross 1 block operator matrix and. then gi...
متن کاملOptimal Portfolio Selection with Singular Covariance Matrix
In this paper we use the Moore-Penrose inverse in the case of a close to singular and ill-conditioned, or singular variance-covariance matrix, in the classic Portfolio Selection Problem. In this way the possible singularity of the variance-covariance matrix is tackled in an efficient way so that the various application of the Problem to benefit from the numerical tractability of the Moore-Penro...
متن کاملThe reverse order law for Moore-Penrose inverses of operators on Hilbert C*-modules
Suppose $T$ and $S$ are Moore-Penrose invertible operators betweenHilbert C*-module. Some necessary and sufficient conditions are given for thereverse order law $(TS)^{ dag} =S^{ dag} T^{ dag}$ to hold.In particular, we show that the equality holds if and only if $Ran(T^{*}TS) subseteq Ran(S)$ and $Ran(SS^{*}T^{*}) subseteq Ran(T^{*}),$ which was studied first by Greville [{it SIAM Rev. 8 (1966...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- J. Multivariate Analysis
دوره 148 شماره
صفحات -
تاریخ انتشار 2016