Convergence acceleration as a dynamical system (Q1339333)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Convergence acceleration as a dynamical system |
scientific article |
Statements
Convergence acceleration as a dynamical system (English)
0 references
5 March 1995
0 references
The dynamical system \(z \to f(z)\) is replaced by \(z \to F_ n(z)\) where \(F_ n\) is given by a general Shanks transformation, i.e. a punctual diagonal Padé approximation leading, in fact, to the quotients of two Hankel or Toeplitz determinants. The particular case of the usual Aitken transformation and its improved form by Steffensen is studied. The author shows that \(F_ n\) retains all finite fixed points of \(f\). Moreover, for a polynomial \(f\) the most poles and zeros of \(F_ n\) accumulate on and inside the Julia set of \(f\) as \(n \to \infty\).
0 references
Steffensen method
0 references
convergence acceleration
0 references
dynamical system
0 references
Shanks transformation
0 references
Padé approximation
0 references
Hankel or Toeplitz determinants
0 references
Aitken transformation
0 references