Quadrature Formulas Based on Rational Interpolation (Q4274396)
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 482004
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Quadrature Formulas Based on Rational Interpolation |
scientific article; zbMATH DE number 482004 |
Statements
Quadrature Formulas Based on Rational Interpolation (English)
0 references
7 July 1994
0 references
quadrature formulas
0 references
rational interpolation
0 references
orthogonal polynomials
0 references
numerical examples
0 references
0.93833494
0 references
0.9245685
0 references
0.92077947
0 references
0.9200046
0 references
0.91586435
0 references
0.91455746
0 references
The authors discuss numerical integration rules that are exact for linear combinations of the basis functions \(1/(1+t_ kx)\). The integration interval is assumed to be \([-1,1]\), with positive weight \(w(x)\). The \(t_ k\)'s are chosen from \((-1,1)\), so that the integration rules should be particularly effective for integrands that have poles outside \([-1,1]\) at the points \(\{-1/t_ k\}\). The authors consider two methods for choosing the integration nodes. The first method uses the \(n\) zeros of the orthogonal function that is found by orthogonalizing the first \(n+1\) basis functions. The second method determines the \(n\) nodes that make the integration rule exact for linear combinations of the first \(2n\) basis functions. The authors show that for each method, the required nodes are zeros of polynomials orthogonal with respect to an appropriately modified weight function, so that in both cases the nodes can be determined using methods that are similar to those used for the classical Gauss rules. The authors also determine asymptotic distribution properties for the nodes, and show that the nodes for different order rules interlace in a manner that is similar to the interlacing of the nodes for the classical Gauss rules. The paper concludes with a report of some results from illustrative numerical examples.
0 references