An analysis of the semilocal convergence for secant-like methods

We present a variation of the technique of proving the semilocal convergence of the family of secant-like methods given in [8], which allows improving the domains of parameters of these methods, along with extending them to situations in which the operator involved is nondifferentiable.