On the rate of convergence of two-term recursions (Q1102474)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 4050202
Language Label Description Also known as
default for all languages
No label defined
    English
    On the rate of convergence of two-term recursions
    scientific article; zbMATH DE number 4050202

      Statements

      On the rate of convergence of two-term recursions (English)
      0 references
      0 references
      0 references
      0 references
      0 references
      1988
      0 references
      Let B be a compact interval in \({\mathbb{R}}\), \(M=B\times B\) and \(\phi\) : \(M\to B\) a map in \(C_ 3(M)\). Suppose that \(\xi\) is a fixed point of \(\phi\). We study the behaviour of the iterates \(x_{n+2}=\phi (x_{n+1},x_ n)\) \((x_ 0,x_ 1\in B)\). Of particular interest is the situation where \(\phi _ x(\xi,\xi)=\phi _ y(\xi,\xi)=0\). In case of the well-known ``Regula falsi'' we also have \(\phi _{xx}(\xi,\xi)=\phi _{yy}(\xi,\xi)=0\) and the order of convergence is \((1+\sqrt{5})\). We consider the case where \(\phi _{yy}(\xi,\xi)\neq 0\). It turns out that there is a constant \(\gamma\) \(\in (1,2)\) such that successive iterates gain factors \(\gamma\),2/\(\gamma\),\(\gamma\),2/\(\gamma\),... on the number of valid decimals. Depending on the initial values \(x_ 0,x_ 1\) the number \(\gamma\) may range over all of (1,2) such that it may occur that an additional iterative step has virtually no effect on the number of correct digits or nearly doubles them.
      0 references
      regula falsi
      0 references
      rate of convergence
      0 references
      two-term recursion
      0 references

      Identifiers