Volume 1 (2) 2007
| Title: | Properties of Preconditioners for Robust Linear Regression |
| Authors: | Venansius Baryamureeba, Trond Steihaug |
| Published: | ©IJCIR Vol1 (2) 2007, PP. 50-66 |
| Language: | English |
Abstract:
In this paper, we consider solving the robust linear regression problem y = Ax + E by an inexact Newton method and an iteratively reweighted least squares method. We show that each of these methods can be combined with the preconditioned conjugate gradient least square algorithm to solve large, sparse systems of linear equations efficiently. We consider the constant preconditioner ATA and preconditioners based on low-rank updates and downdates of existing matrix factorizations. Numerical results are given to demonstrate the effectiveness of these preconditioners. View full Article
| General Terms: | Robust regression, iteratively reweighted least square methods, Newton’s methods, conjugate gradient least squares method, preconditioners Additional Keywords and Phrases: Low rank updates, matrix factorizations |
| Categories and Subject Descriptors: | G.1.0 [Numerical Analysis]: General-Numerical algorithms; G.1.3 [Numerical Analysis]: numerical Linear Algebra—Conditioning; Linear Systems; Spare, structured and very large systems. |
IJCIR Menu
- IJCIR Home
- About IJCIR
- Editorial Board
- Call for Papers
- Submit Papers
- Author Instructions
- Editorial Policies
- Webmail
