Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Cleaning the covariance matrix of strongly nonstationary systems with time-independent eigenvalues

Abstract : We propose a data-driven way to clean covariance matrices in strongly nonstationary systems. Our method rests on long-term averaging of optimal eigenvalues obtained from temporally contiguous covariance matrices, which encodes the average influence of the future on present eigenvalues. This zero-th order approximation outperforms optimal methods designed for stationary systems.
Complete list of metadata

https://hal-centralesupelec.archives-ouvertes.fr/hal-03481441
Contributor : Christian Bongiorno Connect in order to contact the contributor
Submitted on : Wednesday, December 15, 2021 - 12:47:11 PM
Last modification on : Tuesday, December 21, 2021 - 3:45:13 AM

Links full text

Identifiers

  • HAL Id : hal-03481441, version 1
  • ARXIV : 2111.13109

Citation

Christian Bongiorno, Damien Challet, Grégoire Loeper. Cleaning the covariance matrix of strongly nonstationary systems with time-independent eigenvalues. 2021. ⟨hal-03481441⟩

Share

Metrics

Record views

28