EIGENVALUES AND EIGENVECTORS OF SPECTRAL DENSITY MATRICES
Abstract
This report describes some interpretations and uses of eigenvalues and eigenvectors of spectral and sample spectral density matrices of multiple stationary time series. The spectral density matrix of a zero-mean multiple stationary time series is defined. Eigenvalues and eigenvectors of the spectral density matrix are discussed and principal component theory is presented. Statistical distribution theory and related results are used to investigate the eigenvalues of a sample spectral density matrix. This investigation gives methods for obtaining simultaneous confidence bounds on the elements of the true spectral density matrix and its inverse, and also methods for obtaining confidence bounds on the eigenvalues of the true spectral density matrix.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 07, 1967
- Accession Number
- AD0814687
Entities
People
- N. R. Goodman
Organizations
- Teledyne Technologies