论文信息 - Asymptotic analysis of a consistent subspace estimator for observations of increasing dimension

Asymptotic analysis of a consistent subspace estimator for observations of increasing dimension

Traditional estimators of the eigen-subspaces of sample co-variance matrices are known to be consistent only when the sample volume increases for a fixed observation dimension. Due to this fact, their accuracy tends to be rather poor in practical settings where the number of samples and the observation dimension are comparable in magnitude. To overcome this effect, an estimator was recently proposed that provides consistent subspace estimates even when the dimension of the observation scales up with the number of samples. In this paper, the asymptotic distribution of this estimator is characterized by means of a central limit theorem (CLT).

Philippe Loubaton | Walid Hachem | Xavier Mestre | Pascal Vallet

[1] Xavier Mestre,et al. Improved Estimation of Eigenvalues and Eigenvectors of Covariance Matrices Using Their Sample Estimates , 2008, IEEE Transactions on Information Theory.

[2] J. W. Silverstein,et al. No eigenvalues outside the support of the limiting spectral distribution of large-dimensional sample covariance matrices , 1998 .

[3] J. W. Silverstein. Strong convergence of the empirical distribution of eigenvalues of large dimensional random matrices , 1995 .

[4] J. W. Silverstein,et al. COVARIANCE MATRICES , 2022 .

[5] R. Bass,et al. Review: P. Billingsley, Convergence of probability measures , 1971 .

[6] Philippe Loubaton,et al. A New Approach for Mutual Information Analysis of Large Dimensional Multi-Antenna Channels , 2008, IEEE Transactions on Information Theory.