Scholion.ai

Comment on the Iterative Refinement of Least-Squares Solutions

Åke Björck

Journal of the American Statistical Association · 1978 · Volume 73 · Issue 361

DOI: 10.1080/01621459.1978.10480020

Abstract

Three different iterative algorithms for refining a computed least-squares estimate are compared. The first two algorithms are based on the orthogonal decomposition of the matrix X , whereas the third, recently advocated by Fletcher (1975), uses the Cholesky decomposition of X T X. The third algorithm theoretically can give good results even when the condition number κ ( X ) is very large, but in general, the first two algorithms are superior and far more reliable. A variant of one of these may be advantageous when X is large and sparse. Numerical results confirm the theoretical analysis.

Enable JavaScript for citation tools and additional paper interactions.