We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

math.CA

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Mathematics > Classical Analysis and ODEs

Title: Some new perspectives on d-orthogonal polynomials

Abstract: The aim of this paper are two folds. The first part is concerned with the associated and the so-called co-polynomials, i.e., new sequences obtained when finite perturbations of the recurrence coefficients are considered. Moreover, the second part deals with Darboux factorization of Jacobi matrices. Here the respective co-polynomials solutions are explicitly expressed in terms of the fundamental solutions of a (d+2)-term recurrence relation. New identities and formulas related to determinants with co-polynomials entries are obtained. Accordingly, further determinants bring out partial generalizations of Christoffel Darboux formula. Some of new sequences proved useful for determining the entries of matrices in LU and UL decomposition of Jacobi matrix. The last one gives rise of a d-analogue of kernel polynomials with quite a few properties, and further a new characterization of the d-quasi-orthogonality. Kernel polynomials also appear in the (d+1)-decomposition of a d-symmetric sequence. Exploiting properties of d-symmetric sequences, reveal a simple proof of Darboux factorizations. It terms out that Jacobi matrix for d-OPS is a product of d lower bidiagonal matrices and one upper bidiagonal matrix and that each lower bidiagonal matrix is in fact a closed connection between two adjacent components for some d-(symmetric)OPS. Furthermore, we pointed out that if the first component is Hahn classical d-OPS then the corresponding d-symmetric sequence as well as all the components are Hahn classical d-OPS as well. Oscillation matrices assert that zeros of d-OPS are positive and simple whenever the recurrence coefficients are strict positive. Further interlacing properties are justified by the same approach.
Comments: 46 pages. Comments are welcome
Subjects: Classical Analysis and ODEs (math.CA)
Cite as: arXiv:1605.00049 [math.CA]
  (or arXiv:1605.00049v5 [math.CA] for this version)

Submission history

From: Abdessadek Saib [view email]
[v1] Sat, 30 Apr 2016 01:34:14 GMT (46kb)
[v2] Sat, 13 Aug 2016 21:23:19 GMT (45kb)
[v3] Sat, 1 Oct 2016 17:36:19 GMT (46kb)
[v4] Sun, 23 Jul 2017 17:39:23 GMT (47kb)
[v5] Tue, 2 Jan 2018 16:32:22 GMT (47kb)
[v6] Sun, 3 Jan 2021 22:55:22 GMT (49kb)

Link back to: arXiv, form interface, contact.