Recover the spectrum of covariance matrix: a non-asymptotic iterative method
arXiv:2201.00230v1
Originality Incremental advance
AI Analysis
This addresses a fundamental issue in statistical estimation for researchers dealing with covariance matrices, though it appears incremental as it builds on known bias phenomena like the Marchenko-Pastur law.
The authors tackled the problem of bias in the spectrum of sample covariance matrices, introducing an iterative algorithm called 'Concent' that recovers the true spectrum for small and moderate dimensions.
It is well known the sample covariance has a consistent bias in the spectrum, for example spectrum of Wishart matrix follows the Marchenko-Pastur law. We in this work introduce an iterative algorithm 'Concent' that actively eliminate this bias and recover the true spectrum for small and moderate dimensions.