Stieltjes transforms of a class of probability measures

Orthogonal polynomials and functions of hypergeometric type (Jacobi, Laguerre, Hermite, Askey scheme, etc.) (33C45) Integral transforms of special functions (44A20) Probability measures on topological spaces (60B05)

Review text

Let the probability measures \(\mu_N\), \(N= 2,3,\dots\) be defined by \(\mu_N(\{\lambda_{k,N}\})= 1/N\), \(\mu_N(A)= 0\) for \(\lambda_{k,N}\in A\), where \(\lambda_{k,N}\), \(k= 1,2,\dots, N\) are the zeros of the orthogonal polynomial \(P_{N+1}(x)\), which is obtained recursively by \(P_0(x)= 0\), \(P_1(x)= 1\), \[ a_n P_{n+1}(x)+ a_{n-1} P_{n-1}(x)+ b_n P_n(x)= x P_n(x). \] In this paper conditions on \(a_n\) and \(b_n\) are found such that the sequence \(\mu_N\), \(N= 2,3,\dots\) converges weakly to the Dirac measure at the point zero. This is achieved through the convergence of the sequence of Stieltjes transform \(\int^\infty_{-\infty} d\mu_N(t)/(\lambda- t)\) to the function \(1/\lambda\).

Recommendations

Cites work

Cited in

(10)

This page was built for publication: Stieltjes transforms of a class of probability measures

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1874162)