A class of scaled direct methods for linear systems (Q751174)

This paper extends the iterative direct method of the authors and \textit{C. Broyden} [Numer. Math. 45, 361-376 (1984; Zbl 0535.65009)] to include scaling of the system matrix \(A_{m\times n}\). Assuming exact arithmetic, a starting vector is updated in m steps to give the solution x of \(Ax=b\). Each update takes 6 steps which can be chosen such that the method mimics the Gauss-Cholesky factorization, the conjugate gradient method and others.