Acceleration methods for vector sequences (Q1184134)
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: Acceleration methods for vector sequences |
scientific article; zbMATH DE number 34150
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Acceleration methods for vector sequences |
scientific article; zbMATH DE number 34150 |
Statements
Acceleration methods for vector sequences (English)
0 references
28 June 1992
0 references
The purpose of this paper is to propose acceleration methods for certain logarithmic vector sequences. These methods are obtained by extending acceleration methods for scalar logarithmic sequences: Lubkin's \(W\) transform, the \(\theta\)-algorithm and the \(\rho\)-algorithm. Such methods accelerate certain logarithmically convergent vector sequences. However, the vector \(\delta\)-algorithm and the minimal polynomial extrapolation scarcely accelerate such vector sequences. To the author's opinion, all of the \(W\) transforms and the \(\theta\)- algorithms work well on not only certain logarithmically convergent vector sequences, but also on linearly convergent vector sequences. Numerical results showing that acceleration methods are useful for finding zeros of singular equations are given. Extrapolation theorems and error analysis of the new acceleration methods represent future research topics of the author.
0 references
convergence acceleration
0 references
Newton's method
0 references
epsilon algorithm
0 references
rho algorithm
0 references
theta algorithm
0 references
logarithmic vector sequences
0 references
Lubkin's \(W\) transform
0 references
minimal polynomial extrapolation
0 references
Numerical results
0 references
singular equations
0 references
error analysis
0 references
0 references
0 references
0 references