Conditioning analysis of positive definite matrices by approximate factorizations
β Scribed by Robert Beauwens; Renaud Wilmet
- Publisher
- Elsevier Science
- Year
- 1989
- Tongue
- English
- Weight
- 801 KB
- Volume
- 26
- Category
- Article
- ISSN
- 0377-0427
No coin nor oath required. For personal study only.
π SIMILAR VOLUMES
A Newton-Kantorovich algorithm used in the approximation of symmetrically reciprocal positive matrices by transitive matrices is found to be more generally useful in the approximation of merely positive matrices. The occurrence of multiple stationary values when the approximated matrix has certain s
We study the geometrical properties of the Frobenius condition number on the cone of symmetric and positive definite matrices. This number, related to the cosine of the angle between a given matrix and its inverse, is equivalent to the classical 2-norm condition number, but has a direct and natural
In this paper a new ILU factorization preconditioner for solving large sparse linear systems by iterative methods is presented. The factorization which is based on Abiorthogonalization process is well defined for a general positive definite matrix. Numerical experiments illustrating the performance