Recovery Limits for Eigenvector Alignment in HDLSS Spiked Covariance Models
En palabras de los autores
In a spiked covariance model, let and be selected sample and population eigenvector matrices. Their alignment matrix is a latent oracle diagnostic of direction-specific correspondence beyond subspace overlap. We study its uniform pathwise recovery in a nested high-dimensional, low-sample-size experiment with fixed sample size and growing dimension. At every fixed model, an observable spectral contrast consistently estimates the row Gram matrix . For two or more spikes, however, no measurable estimator sequence achieves uniformly reliable almost-sure path recovery of over a family having common spike strengths and a common signal subspace but different ordered eigenframes. For every , a finite orbit forces each estimator to have some orbit point at which the approximate path-recovery probability is at most a variational bound , where tends to zero as tends to zero. The worst-model exact-recovery probability over a countable orbit is zero. Thus the Gram summary is recoverable model by model, while the full alignment matrix is not uniformly recoverable along the nested path.
Apareció: martes, 22 de septiembre. arXiv. Preprint, todavía sin revisión por pares.
Comentario de los autores: 24 pages. Presented at the SIAM Conference on Financial Mathematics and Engineering (FM25), Florida, 2025