ORTHOGONAL LINEAR-REGRESSION ALGORITHM-BASED ON AUGMENTED MATRIX FORMULATION

Citation
A. Bargiela et Jk. Hartley, ORTHOGONAL LINEAR-REGRESSION ALGORITHM-BASED ON AUGMENTED MATRIX FORMULATION, Computers & operations research, 20(8), 1993, pp. 829-836
Citations number
2
Categorie Soggetti
Operatione Research & Management Science","Computer Applications & Cybernetics","Operatione Research & Management Science
ISSN journal
03050548
Volume
20
Issue
8
Year of publication
1993
Pages
829 - 836
Database
ISI
SICI code
0305-0548(1993)20:8<829:OLAOAM>2.0.ZU;2-L
Abstract
The problem of n-dimensional orthogonal linear regression is a problem of finding an n-dimensional hyperplane minimising the sum of Euclidea n distances between this hyperplane and a given set of m points, where m greater-than-or-equal-to n. This nonlinear programming problem has been re-cast in an augmented matrix form and solved as a sequence of i teratively re-weighted least square problems. The proposed algorithm i s seen as an alternative to the recently published algorithm by Cavali er and Melloy [1].