Multivariate Jacobi polynomials and the Selberg integral. II (Q2628947)

From MaRDI portal
Revision as of 07:15, 12 July 2024 by ReferenceBot (talk | contribs) (‎Changed an Item)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
scientific article
Language Label Description Also known as
English
Multivariate Jacobi polynomials and the Selberg integral. II
scientific article

    Statements

    Multivariate Jacobi polynomials and the Selberg integral. II (English)
    0 references
    19 July 2016
    0 references
    This paper is a continuation of the authors' article [Funct. Anal. Appl. 46, No. 4, 262--278 (2012; Zbl 1306.43006); translation from Funkts. Anal. Prilozh. 46, No. 4, 31--50 (2012)]. Let \(H_N\) be the Hilbert space of all functions on \([-1,1]^N\) that are invariant under the permutation of coordinates and are square integrable with respect to the measure \(m^{a,b,\theta}_N\) on \([-1,1]^N\) which has the density \(\prod_{i=1}^N(1-x_i)^a(1+x_i)^bV^{2\theta}(x)\) relative to the Lebesgue measure, where \(V(x)\) is the Vandermonde determinant. Let \(\mathcal V_N\) be the set of \(N\)-tuples of nonincreasing nonnegative integers. The \(N\)-variate Jacobi polynomials with parameters \((a,b,\theta)\) are symmetric polynomials \(\pi^{a,b,\theta}_\lambda(x_1,\dotsc,x_N)\) indexed by arbitrary \(N\)-tuples \(\lambda\in\mathcal V_N\). The stochastic matrix \(\Lambda^N_{N-1}(\lambda,\nu)\) is defined from the equality \[ \Phi^{a,b,\theta}_\lambda(x_1,\dotsc,x_{N-1},1)= \sum_{\nu\in\mathcal V_{N-1}}\Lambda^N_{N-1}(\lambda,\nu) \Phi^{a,b,\theta}_\nu(x_1, \dotsc,x_{N-1}), \] where \(\Phi^{a,b,\theta}_\lambda(x_1,\dotsc,x_{N})=\pi^{a,b,\theta}_\lambda(x_1,\dotsc,x_{N})/\pi^{a,b,\theta}_\lambda(1,\dotsc,1)\). If \(\{M_N:N=1,2,\dotsc\}\) is a family of complex measures with finite variation on the set \(\mathcal V_N\) then the family \(\{M_N\}\) is called coherent if \(M_N\Lambda^N_{N-1}=M_{N-1}\) for every \(N\geq2\). At last the \(z\)-measure on \(\mathcal V_N\) with parameters \((z,z')\in D\times D\) is called the complex measure given by \[ M_N(\lambda|z,z',a,b,\theta)=\frac{(f_{z|N},\pi^{a,b,\theta}_\lambda)(\pi^{a,b,\theta}_\lambda,f_{\bar z'|N})} {(f_{z|N},f_{\bar z'|N}) \|\pi^{a,b,\theta}_\lambda)\|^2}, \] where \(D=\{z\in\mathbb C:\text{Re}z>-(1+b)/2\}\) and \(f_{z|N}=\prod_{i=1}^N(1+x_i)^z\in H_N\). The main result of the paper is (Theorem 4.1): Assume that the parameters \(z\) and \(z'\) lie in the half-plane \(D\), then the family of \(z\)-measures \(\{M_N(\lambda|z,z',a,b, \theta)\}\) is coherent.
    0 references
    multivariate Jacobi polynomials
    0 references
    symmetric polynomials
    0 references
    Lebesgue measure
    0 references
    stochastic matrix
    0 references
    complex measures
    0 references
    Selberg integral
    0 references
    0 references
    0 references

    Identifiers