Markov-type inequalities for polynomials with restricted zeros

From MaRDI portal
(Redirected from Publication:1960906)





Let \(D^{\alpha}\) be a lens shaped, symmetric with respect to the real axis region in \(\mathbb C\), bounded by two circular arcs joining 1 and -1 and meeting each other at an inner angle \(\alpha\pi\) in \(\pm 1\). The author shows the following version of Markov's inequality for polynomials with restrictions on the distribution of the zeros: For every \(0\leq\alpha<1\) there exists a constant \(c(\alpha)\) depending only on \(\alpha\) such that \[ \|p^\prime\|\leq c(\alpha)n^{2-\alpha}\|p\| \] for polynomials \(p\) of degree \(n\) nonvanishing on \(D^\alpha\). Here \(\|\cdot\|=\sup|\cdot|([-1,1])\). Applying this result to the polynomial \(q(z)=\sup_{z\in D^\alpha}|p(z)|+p(z)\) yields, in particular, the well-known Szegö estimate \[ |p^\prime(\pm 1)|\leq c(\alpha)n^{2-\alpha}\sup_{z\in D^\alpha}|p(z)|. \]











This page was built for publication: Markov-type inequalities for polynomials with restricted zeros

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1960906)