Convergence rates for inexact Newton-like methods at singular points and applications (Q1294207)

This paper deals with inexact Newton-like methods of the form \(x_{n+1}=x_n-T(x_n)^{-1}F(x_n)-v_n\), \(v_n = v(x_n)\), where \(T(x)\in L(E_1,E_2)\), \(x\in E_1\) is an approximation to the Fréchet derivative \(F'(x)\) of operator \(F\) for all \(x\in E_1\), \(v:E_1\to E_1\) is a continuous operator, and \(f\) is an operator defined on a Banach space \(E_1\) with values in a Banach space \(E_2\). The points \(v_n\) are determined in such way that the iteration \(\{x_n\}\) converges to a solution \(x^{\ast}\) of an equation \(F(x)=0\). The author investigates the case when \(T(x^{\ast})\) is a Fredholm operator of index zero. Convergence rates are provided for studying inexact Newton-likes methods. Numerical results are presented for some examples appearing in connection with some predator-prey problems appearing in population modelling.