The impact of equal weighting of low- and high-confidence observations on robust linear regression computations

BIT Numerical Mathematics

Volumn 41, Issue 5, 2001, Pages 847-855

  Baryamureeba, Venansius ^a  

^a MAKERERE UNIVERSITY   (Uganda)

Author keywords

Basis identification techniques; Conjugate gradient least squares method; LSQR; Newton's method; Preconditioner; Robust linear regression

Indexed keywords

EID: 27844557155     PISSN: 00063835     EISSN: None     Source Type: Journal    
DOI: 10.1023/A:1021912522498     Document Type: Article

References (4)

1. M. Adlers, Topics in Sparse Least Squares Problems, PhD thesis, Department of Mathematics, Linköping University, Sweden, 2000.
2. J. Antoch and H. Ekblom, Recursive robust regression computation aspects and comparison, Computational Statistics and Data Analysis, 19 (1995), pp. 115-128.
3. V. Baryamureeba, Solution of robust linear regression problems by Krylov subspace methods, Technical Report No. 203, Department of Informatics, University of Bergen, Bergen, Norway.
4. V. Baryamureeba, Solution of large-scale weighted least squares problems, Numer. Linear Algebra Appl., to appear.