A remark on the inverse problem for orthogonal polynomials (Q1584103)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | A remark on the inverse problem for orthogonal polynomials |
scientific article |
Statements
A remark on the inverse problem for orthogonal polynomials (English)
0 references
31 October 2000
0 references
Let \(x_n\) be the maximal root of the orthogonal polynomial and \(y_n\) be the root next to the maximal one; \(\widehat\mu(z)=\int_{i\infty}^\infty \frac{d\mu(x)}{z-x}\sim \sum_{k=0}^\infty \frac{\mu_k}{z^{k+1}}\), \(\mu\) is the measure on \(R\). From [\textit{A.~A.~Gonchar}, Approximation Theory V, Proc. 5-th Int. Symp., College Station/Tex. 1986, 55-70 (1986; Zbl 0645.41014)]\ it is known that, if (1) \(\overline\lim|\widehat x-x_n|^{1/n}<q<1\), (2) \(y_n\) are bounded away from \(\widehat x\), then the function \(\widehat\mu(t)\) has a pole (of the first order) in point \(\widehat x\), and therefore, the measure \(\mu\) has an isolated mass in this point. The author shows that for the polynomials being orthogonal with respect to the symmetric measures (i.e. \(\mu(E)=\mu(-E)\) for any Borel set \(E\subset R)\) the condition (2) can be weakened.
0 references
Nevanlinna-peak problem
0 references
orthogonal polynomial zeros
0 references