Robust Estimates of Covariance Matrices in Large Dimensional Regime

SIRV, covariance matrix estimates, statistical performance analysis, bias, consistency, asymptotic distribution, non-Gaussian noise.. Forster is with the Groupe

This is in particular the case in financial data analysis where few stationary monthly observations of numerous stock indexes are used to estimate the joint covariance matrix of

The contribution to the expectation (12) of paths with no 1-edges has been evaluated in Proposition 2.2. This section is devoted to the estimation of the contribution from edge paths

We establish a large deviation principle for the empirical spec- tral measure of a sample covariance matrix with sub-Gaussian entries, which extends Bordenave and Caputo’s result

Based on these evidentiary support and motivated by the fact that the eigenmatrix of Wishart matrix is Haar (uniformly) distributed, we believe that the eigenmatrix of a

Moreover re- marking some strong concentration inequalities satisfied by these matrices we are able, using the ideas developed in the proof of the main theorem, to have some

In Section 2 we state the assumptions and main results of the article: Proposition 2.1 and Theorem 2.2 are devoted to the limiting behaviour and fluctuations of λ max (S N );

Under a mild dependence condition that is easily verifiable in many situations, we derive that the limiting spectral distribution of the associated sample covariance matrix