A VECTOR SPACE DERIVATION-USING DYADS-OF WEIGHTED LEAST SQUARES FOR CORRELATED NOISE
Abstract
Matrix-analysis and recursive matrix computing subroutines offer hope of relieving the current computer data deluge. Classical weighted least squares for multivariable parameter estimation in the presence of correlated noise are developed in a geometrical vector space setting. Rank-one matrices, or dyads, are used extensively, especially in obtaining gradients of traces of variance matrices.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1968
- Accession Number
- AD0674190
Entities
People
- James S. Pappas.
Organizations
- United States Army Test and Evaluation Command