Improbability of nonconvergence in a cubic root-finding method (Q1107954): Difference between revisions

From MaRDI portal
Import240304020342 (talk | contribs)
Set profile property.
ReferenceBot (talk | contribs)
Changed an Item
 
Property / cites work
 
Property / cites work: Q5584339 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q3906149 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q3324927 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Newton’s Algorithm and Chaotic Dynamical Systems / rank
 
Normal rank
Property / cites work
 
Property / cites work: Newton's Method, Circle Maps, and Chaotic Motion / rank
 
Normal rank
Property / cites work
 
Property / cites work: Computational complexity. On the geometry of polynomials and a theory of cost. I / rank
 
Normal rank
Property / cites work
 
Property / cites work: On inequalities of the Turán type / rank
 
Normal rank
Property / cites work
 
Property / cites work: Improbability of nonconvergent chaos in Newton's method / rank
 
Normal rank
Property / cites work
 
Property / cites work: Newton's Method and Symbolic Dynamics / rank
 
Normal rank

Latest revision as of 18:45, 18 June 2024

scientific article
Language Label Description Also known as
English
Improbability of nonconvergence in a cubic root-finding method
scientific article

    Statements

    Improbability of nonconvergence in a cubic root-finding method (English)
    0 references
    1988
    0 references
    This paper is concerned with the structure of the nonconvergent set NC(Ef) of initial points for which the Euler method for finding the roots of real functions \[ x_{k+1}=Ef(x_ k),\quad where\quad Ef(x)=x- [f(x)/f'(x)][1+f(x)\cdot f''(x)/2(f'(x))^ 2] \] fails to converge. The theorem is proved, that NC(Ef) is a Cantor set of Lebesgue measure zero. As referred in the paper, the conditions of the theorem (they are satisfied, for example, if f is a polynomial with all roots real) are adapted from similar conditions used by \textit{D. Saari} and \textit{J. Urenko} [Am. Math. Mon. 91, 3-17 (1984; Zbl 0532.58016)] and \textit{J. Urenko} [J. Math. Anal. Appl. 117, 42-47 (1986; Zbl 0634.58008)], where a similar result was shown for Newton's method. The theorem is established by reducing questions about the Euler method to a problem on Newton's method.
    0 references
    0 references
    root-finding method
    0 references
    nonconvergent set
    0 references
    Euler method
    0 references
    roots of real functions
    0 references
    Cantor set
    0 references
    0 references