Chebyshev solution of large linear systems
β Scribed by J.B. Rosen
- Publisher
- Elsevier Science
- Year
- 1967
- Tongue
- English
- Weight
- 636 KB
- Volume
- 1
- Category
- Article
- ISSN
- 0022-0000
No coin nor oath required. For personal study only.
β¦ Synopsis
The general problem considered is that of solving a linear system of equations which is singular or almost singular. A method is described which obtains a "solution" to the system which is stable with respect to small changes in the matrix elements. This method will solve an overdetermined system in m variables and n equations (m < n) even when the system rank is less than m, and should therefore be very useful in many statistical applications. In this case the error of the system is minimized in the Chebyshev norm using a linear programming formulation and solution. A numerical example using the Hilbert matrix is described in detail.
π SIMILAR VOLUMES
We developed a direct out-of-core solver for dense non-symmetric linear systems of 'arbitrary' size N;N. The algorithm fully employs the Basic Linear Algebra Subprograms (BLAS), and can therefore easily be adapted to different computer architectures by using the corresponding optimized routines. We