Quantitative estimates of the convergence of the empirical covariance matrix in log-concave ensembles

@article{Adamczak2009QuantitativeEO,
  title={Quantitative estimates of the convergence of the empirical covariance matrix in log-concave ensembles},
  author={Radosław Adamczak and Alexander E. Litvak and Alain Pajor and Nicole Tomczak-Jaegermann},
  journal={Journal of the American Mathematical Society},
  year={2009},
  volume={23},
  pages={535-561}
}
Let K be an isotropic convex body in Rn. Given e > 0, how many independent points Xi uniformly distributed on K are neededfor the empirical covariance matrix to approximate the identity up to e with overwhelming probability? Our paper answers this question from [12]. More precisely, let X ∈ Rn be a centered random vector with a log-concave distribution and with the identity as covariance matrix. An example of such a vector X is a random point in an isotropic convex body. We show that for any e… 
View PDF on arXiv
174 Citations
Highly Influential Citations
25
Background Citations
73
Methods Citations
27
Results Citations
6

174 Citations

On the convergence of the extremal eigenvalues of empirical covariance matrices with dependence

Consider a sample of a centered random vector with unit covariance matrix. We show that under certain regularity assumptions, and up to a natural scaling, the smallest and the largest eigenvalues of…

Restricted isometry property for random matrices with heavy-tailed columns

Random embeddings with an almost Gaussian distortion

Invertibility via distance for noncentered random matrices with continuous distributions

TLDR
The method is principally different from a standard approach involving a decomposition of the unit sphere and coverings, as well as an approach of Sankar-Spielman-Teng for non-centered Gaussian matrices.

Covariance estimation for distributions with 2+ε moments

We study the minimal sample size N=N(n) that suffices to estimate the covariance matrix of an n-dimensional distribution by the sample covariance matrix in the operator norm, with an arbitrary fixed…

How Close is the Sample Covariance Matrix to the Actual Covariance Matrix?

Given a probability distribution in ℝn with general (nonwhite) covariance, a classical estimator of the covariance matrix is the sample covariance matrix obtained from a sample of N independent…

On a multi-integral norm defined by weighted sums of log-concave random vectors

Let C and K be centrally symmetric convex bodies in R n . We show that if C is isotropic then for every s > 1 and t = ( t 1 , . . . , t s ) ∈ R s , where L C is the isotropic constant of C and M ( K…

COVARIANCE ESTIMATION FOR DISTRIBUTIONS WITH 2+epsilon MOMENTS

We study the minimal sample size N = N ( n ) that suffices to estimate the covariance matrix of an n -dimensional distribution by the sample covariance matrix in the operator norm, with an arbitrary…

Almost sure convergence of the largest and smallest eigenvalues of high-dimensional sample correlation matrices

A pr 2 01 8 Approximating the covariance ellipsoid

We explore ways in which the covariance ellipsoid B = {v ∈ R : E 〈X, v〉2 ≤ 1} of a centred random vector X in R can be approximated by a simple set. The data one is given for constructing the…
...

References

SHOWING 1-10 OF 36 REFERENCES

Lp-moments of random vectors via majorizing measures

On the limiting empirical measure of eigenvalues of the sum of rank one matrices with log-concave distribution

We consider n × n real symmetric and hermitian random matrices Hn that are sums of a non-random matrix H n and of mn rank-one matrices determined by i.i.d. isotropic random vectors with log-concave…

Reconstruction and subgaussian operators

We present a randomized method to approximate any vector v from some set T ⊂ R n . The data one is given is the set T and k scalar products (h Xi, vi ) k=1 , where (Xi) k=1 are i.i.d. isotropic…

Restricted Isometry Property of Matrices with Independent Columns and Neighborly Polytopes by Random Sampling

This paper considers compressed sensing matrices and neighborliness of a centrally symmetric convex polytope generated by vectors ±X1,…,±XN∈ℝn, (N≥n). We introduce a class of random sampling matrices…

Random Vectors in the Isotropic Position

Abstract Letybe a random vector in R n, satisfying E y⊗y=id. LetMbe a natural number and lety1, …, yMbe independent copies ofy. We study the question of approximation of the identity operator by…

Random Points in Isotropic Convex Sets

Let K be a symmetric convex body of volume 1 whose inertia tensor is isotropic, i.e., for some constant L we have R K〈x, y〉2 dx = L2|y|2 for all y. It is shown that if m is about n(log n)3 then with…

On the Limiting Empirical Measure of the sum of rank one matrices with log-concave distribution

We consider $n\times n$ real symmetric and hermitian random matrices $H_{n,m}$ equals the sum of a non-random matrix $H_{n}^{(0)}$ matrix and the sum of $m$ rank-one matrices determined by $m$ i.i.d.…

On weakly bounded empirical processes

Let F be a class of functions on a probability space (Ω, μ) and let X1,...,Xk be independent random variables distributed according to μ. We establish an upper bound that holds with high probability…

Sampling convex bodies: a random matrix approach

We prove the following result: for any e > 0, only C(e)n sample points are enough to obtain (1 + e)-approximation of the inertia ellipsoid of an unconditional convex body in R". Moreover, for any p >…

On singular values of matrices with independent rows

We present deviation inequalities of random operators of the form 1 N ∑N i=1 Xi ⊗ Xi from the average operator E(X ⊗ X), where Xi are independent random vectors distributed as X, which is a random…