Better together: Cross and joint covariances enhance signal detectability in undersampled data
- 17 September 2026
- journal article
- research article
- Published by American Physical Society (APS) in Physical Review E
- Vol. 114 (3) , 035305
Abstract
Random matrix theory reveals that analyzing joint or cross-covariance matrices detects shared signals between high-dimensional variables earlier than standard individual self-covariances under heavy sampling noise.Keywords
All Related Versions
Funding Information
- National Science Foundation (DMS-2235451)
- Simons Foundation (MPS-NITMB-00005320)
This publication has 37 references indexed in Scilit:
- Spiked multiplicative random matrices and principal componentsStochastic Processes and their Applications, 2023
- Large-scale neural recordings with single neuron resolution using Neuropixels probes in human cortexNature Neuroscience, 2022
- Large-scale neural recordings call for new insights to link brain and behaviorNature Neuroscience, 2022
- A First Course in Random Matrix Theory2020
- On the principal components of sample covariance matricesProbability Theory and Related Fields, 2015
- Automated image-based tracking and its application in ecologyTrends in Ecology & Evolution, 2014
- The singular values and vectors of low rank perturbations of large rectangular random matricesJournal of Multivariate Analysis, 2012
- The eigenvalues and eigenvectors of finite, low rank perturbations of large random matricesAdvances in Mathematics, 2011
- Spatial Pattern Analysis of Functional Brain Images Using Partial Least SquaresNeuroImage, 1996
- Analysis of a complex of statistical variables into principal components.Journal of Educational Psychology, 1933