CMV matrices and little and big \(-1\) Jacobi polynomials (Q1935030)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | CMV matrices and little and big \(-1\) Jacobi polynomials |
scientific article |
Statements
CMV matrices and little and big \(-1\) Jacobi polynomials (English)
0 references
30 January 2013
0 references
The authors introduce a new map from polynomials orthogonal on the unit circle (OPUC) to the real line \(\mathbb{R}\). This map may be interpreted as a one-parameter generalization of the Delsarte-Genin map from OPUC to symmetric orthogonal polynomials on intervals. This new map is derived using elementary properties of the CMV polynomials (which are the unitary analogs of the Jacobi matrices). Starting from two such unitary elementary matrices \(L\) and \(M\), a 3-diagonal Jacobi matrix is defined by \(J = L + M\). The new family \(Q_n (x)\) of orthogonal polynomials on \(\mathbb{R}\) is then associated with the matrix \(J\). It is shown that the polynomials \(Q_n(x)\) coincide with the Delsarte-Genin polynomials \(\widetilde{P}_n(x)\), which are obtained from symmetric orthogonal polynomials \(S_n(x)\) via a Christoffel transform, and in the case of the Jacobi OPUC with the little \(-1\)-Jacobi polynomials. The authors then generalize their approach by considering a one-parameter family of matrices \(K(\lambda) : = L + \lambda M\), \(\lambda\in \mathbb{R}\). The matrices \(K(\lambda)\) generate new families of orthogonal polynomials \(Q_n(x;\lambda)\). It is proved that, in the case of the Jacobi OPUC, the polynomials \(Q_n(x;\lambda)\) coincide with the big \(-1\)-Jacobi polynomials which are orthogonal on the union of two intervals in \(\mathbb{R}\).
0 references
orthogonal polynomials on the unit circle
0 references
little and big \(-1\)-Jacobi polynomials
0 references
Delsarte-Genin map and its generalization
0 references
CMV matrix
0 references
Jacobi matrix
0 references
linear pencils
0 references
0 references
0 references
0 references