The uniform convergence of subsequences of the last intermediate row of the Padé table (Q1395803): Difference between revisions
From MaRDI portal
Removed claim: reviewed by (P1447): Item:Q476372 |
Changed an Item |
||
Property / reviewed by | |||
Property / reviewed by: Alexander V. Tovstolis / rank | |||
Normal rank |
Revision as of 20:31, 15 February 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | The uniform convergence of subsequences of the last intermediate row of the Padé table |
scientific article |
Statements
The uniform convergence of subsequences of the last intermediate row of the Padé table (English)
0 references
1 July 2003
0 references
The article is devoted to the investigation of the uniform convergence of rows of the Padé approximants \(\pi_{n,m}(z)\) for a meromorphic function \(a(z)\). As the author notes, the complete description of the limit point set of the poles of \(\pi_{n,m}\) as \(n\to\infty\) does not exists for any row. In the article the author constructs the complete theory of the uniform convergence of the sequence \(\pi_{n,\lambda-1}(z)\), i.e., for the case of last intermediate row. Here \(\lambda\) is the number of the poles (with regard to multiplicity) of \(a(z)\) in the disk \(|z|<R\). The author investigates the limits of all convergent subsequences of the sequence of the denominators \(Q_{n,\lambda-1}(z)\) of \(\pi_{n,\lambda-1}(z)\) as \(n\to\infty\). Thus, for this row one can explicitly describe the set of limit points of the poles of \(\pi_{n,\lambda-1}(z)\). It is established that the asymptotic behavior of the denominators \(Q_{n,\lambda-1}(z)\) is defined by an arithmetic nature of the dominant poles of the function \(a(z)\), i.e., the poles that have maximal modulus and maximal multiplicity. The limits of all convergent subsequences of \(\{Q_{n,\lambda-1}(z)\}\) form a family of polynomials that is parametrized by a monothetic subgroup \({\mathbb F}\) of the torus \({\mathbb T}^\nu\). The group \({\mathbb F}\) is constructed using the arguments \(\Theta_1,\dots,\Theta_\nu\) of dominant poles of \(a(z)\). The author's approach to a convergence problem for a row sequence of Padé approximants is based on a method of essential polynomials introduced in the recent author's article [Factorization of triangular \(2\times 2\) matrix functions. Manuscript 5930-82, deposited at VINITI (Russian) (1982)]. The present article is a serious investigation. A lot of original methods are developed and applied to obtain the main results. It may be very useful for the specialists in approximation theory. The article contains an extended bibliography that reflects several results obtained in this area.
0 references
Padé approximation
0 references
Padé table
0 references
meromorphic function
0 references
dominant pole
0 references
set of additional limit points
0 references
essential polynomials
0 references
Wiener-Hopf factorization
0 references
asymptotic behavior
0 references
parameter group
0 references