On the asymptotic behaviour of the eigenvalue distribution of block correlation matrices of high-dimensional time series
arXiv:2004.07226
Abstract
We consider linear spectral statistics built from the block-normalized correlation matrix of a set of mutually independent scalar time series. This matrix is composed of blocks that contain the sample cross correlation between pairs of time series. In particular, each block has size and contains the sample cross-correlation measured at consecutive time lags between each pair of time series. Let denote the total number of consecutively observed windows that are used to estimate these correlation matrices. We analyze the asymptotic regime where while , . We study the behavior of linear statistics of the eigenvalues of this block correlation matrix under these asymptotic conditions and show that the empirical eigenvalue distribution converges to a Marcenko-Pastur distribution. Our results are potentially useful in order to address the problem of testing whether a large number of time series are uncorrelated or not.
References in corpus (7)
- Averaging Fluctuations in Resolvents of Random Band Matrices
- Covariance matrix estimation for stationary time series
- Convergence of the largest singular value of a polynomial in independent Wigner matrices
- Gaussian fluctuations for linear spectral statistics of large random covariance matrices
- Large sample behaviour of high dimensional autocovariance matrices
- Spectra of large time-lagged correlation matrices from Random Matrix Theory
- Large random matrix approach for testing independence of a large number of Gaussian time series