RUS  ENG
Full version
JOURNALS // Matematicheskoe modelirovanie // Archive

Matem. Mod., 2012 Volume 24, Number 3, Pages 113–136 (Mi mm3273)

This article is cited in 3 papers

On the Craig method convergency for linear algebraic systems

N. N. Kalitkin, L. V. Kuzmina

Keldysh Institute of Applied Mathematics of Rus. Acad. Sci., Moscow

Abstract: The iterative Craig method permits to solve linear algebraic systems with nonsymmetric (and even rectangular) matrix. The simple form of this method was constracted. The convergention this method was inverstigated on tests. The comparison with the conjugated gradients method was fulfeeld. It occurred that round of errors for the Craig method decelerate essentially iterations convergence, but not prevent from high accuracy achievement (for well conditioned matrixes). The effective criterium is found for iterations truncation.

Keywords: linear algebraic systems, the Craig method, iterations convergency, round of errors.

Received: 14.04.2011


 English version:
Mathematical Models and Computer Simulations, 2012, 4:5, 509–526

Bibliographic databases:


© Steklov Math. Inst. of RAS, 2024