Interval iteration for zeros of systems of equations (Q1064017)
From MaRDI portal
| This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Interval iteration for zeros of systems of equations |
scientific article; zbMATH DE number 3919666
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Interval iteration for zeros of systems of equations |
scientific article; zbMATH DE number 3919666 |
Statements
Interval iteration for zeros of systems of equations (English)
0 references
1985
0 references
Let \(F: D\subseteq {\mathbb{R}}^ n\to {\mathbb{R}}^ n\) be a continuously differentiable function such that F'(x) is regular for all \(x\in D\). To find a zero \(x^*\) of F in D, a number of good local methods are available. Such methods are globally convergent only under very restrictive conditions. The author shows that a suitable combination of any good local method with certain techniques from interval mathematics produces a strategy globally convergent under easily verifiable conditions. In particular, a quadratically convergent method is described. If the selected initial region contains no zero of F, this fact is detected after finitely many iterations. The use of rounded interval arithmetic provides rigorous (and small!) error bounds for the computed approximation to \(x^*\).
0 references
global convergence
0 references
local method
0 references
interval mathematics
0 references
quadratically convergent method
0 references
rounded interval arithmetic
0 references
error bounds
0 references