Local convergence of a secant type method for solving least squares problems (Q618062): Difference between revisions

The problem of approximating a solution of the nonlinear least square problem is considered. This problem can be solved by Gauss-Newton-type methods. A new local convergence analysis for the iterative method, alternative to Gauss-Newton type methods is provided. This method uses the special choice of a linear operator which is a divided difference or a linear combination of divided differences. The radius of convergence is determined as well as usable error estimates. Numerical examples are also provided.