The diagonalisation of the multivariate Bernstein operator (Q1867492): Difference between revisions
From MaRDI portal
Changed an Item |
Set profile property. |
||
Property / MaRDI profile type | |||
Property / MaRDI profile type: MaRDI publication profile / rank | |||
Normal rank |
Revision as of 05:00, 5 March 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | The diagonalisation of the multivariate Bernstein operator |
scientific article |
Statements
The diagonalisation of the multivariate Bernstein operator (English)
0 references
2 April 2003
0 references
The paper is split into five sections. In Section 1 is defined the multivariate Bernstein operator \(B_n\) of degree \(n\) for a simplex in \(\mathbb{R}^s\), establishing notations and some technical results. In Section 2, the authors show that \(B_n\) is diagonalisable with the same eigenvalues as the univariate Bernstein operator, i.e. \(\lambda_k^{(n)}= {n!\over (n-k)!} {1 \over n^k}\) \((k=1,2, \dots,n)\) and \(1=\lambda_1^{(n)} >\lambda_2^{(n)}> \cdots> \lambda_n^{(n)}>0\). Moreover, since \(B_n\) reproduces only the linear polynomials, these are the eigenspace for \(\lambda_1^{(n)} =1\). For \(k>1\), the \(\lambda_k^{(n)}\)-eigenspace consists of polynomials of exact degree \(k\), which are uniquely determined by their leading term. In Section 3, they show that \(B_nf\) takes a simplified form when \(f\) are certain ridge-type functions. This is used to describe the eigenfunctions of \(B_n\) in term of the substitution of the barycentric coordinates (for the underlying simplex) into elementary eigenfunctions. In Section 4 is proved, as in the univariate case, that the eigenfunctions, with fixed leading term, converge as \(n\to\infty\). Furthermore, are described these limiting eigenfunctions and their connection with orthogonal polynomials of several variables. Section 5 contains a result about \(B_n\), applied to certain shifted factorials, and some consequences of it.
0 references
multivariate Bernstein operator
0 references
diagonalization
0 references
eigenvalues
0 references
eigenfunctions
0 references