An improvement of derivative-free point-to-point iterative processes with central divided differences

Handle

https://riunet.upv.es/handle/10251/204097

Cita bibliográfica

Hernández-Verón, MA.; Magreñán, ÁA.; Martínez Molada, E.; Singh, S. (2023). An improvement of derivative-free point-to-point iterative processes with central divided differences. International Journal of Nonlinear Sciences and Numerical Simulation. 24(7):2781-2799. https://doi.org/10.1515/ijnsns-2021-0245

Titulación

Resumen

[EN] In this article, we introduce a new Steffensen-type method with the advantage that its behavior is very similar to Newton's method; therefore, it is a very remarkable way of avoiding the drawback that Newton's method presents for nondifferentiable operators. In our study, we perform an exhaustive comparative study between the semilocal convergence of Newton's method and the derivative-free point-to-point iterative process considered; in the case of differentiable operators, we use the majoring sequences and the majorant principle. In the nondifferentiable case, we impose conditions on the starting point and on the nonlinear operator to obtain a semilocal convergence result for the iterative process considered. In both cases, we complete the theoretical convergence proofs with a dynamical study and a numerical test. In the case of differentiable operators, this study confirms that the accessibility and numerical behavior of both iterative processes, Newton's method and the derivative-free point-to-point iterative process considered, are very similar.

Fuente

International Journal of Nonlinear Sciences and Numerical Simulation issn: 1565-1339

Enlaces relacionados

URL