Directional secant-type methods for solving equations (Q353169)

The article of I. K. Argyros and S. Hilout is a valuable contribution to the calculus of solving equations with a special motivation and application in optimization, especially. This investigation takes place in a wide model setting, and it employs numerical analysis by addressing some of its basic solution techniques in a rather advanced way. This paper is well structured, well written and well shown computationally. Its progress and advantages, in comparison with contributions from literature, are made clear. Via the addressing of solving of equations, this work invites to the consideration of how to use it practically and fruitfully. The authors provide a semilocal convergence analysis on directional secant-type methods for the multidimensional space. Employing weaker hypotheses than the ones used by An and Bai, they provide a semilocal convergence analysis with the subsequent four advantages: weak convergence conditions, larger convergence domain, finer error estimates on the distances involved, and more precise information on the location of the solution. Moreover, a computational example is given, where the authors' results apply to solve an equation but not the ones of An and Bai. In another example, the authors demonstrate the implementation of their method. The six sections of this article are as follows: 1. Introduction, 2. Semilocal convergence analysis of (DNGSM) using recurrent sequences, 3. Semilocal convergence analysis of (DMCMSM), 4. Semilocal Ccnvergence for (DNGSM) and (DMCMSM), 5. Numerical examples, and 6. Concluding remarks. In future, further refinements of numerical methods and strong results could be expected, initialized by this article. Those advances could initiate and support achievements in emerging fields of science, engineering, economics, game theory, finance, OR, medicine and healthcare, and, eventually, to improvements of living conditions on earth.