On the Area of the Image of a Bounded Analytic Function

From MaRDI portal





Let f be analytic in the unit disc, and put \[ A(r,f)=\int^{2\pi}_{0}\int^{r}_{0}| f'(\rho e^{i\theta})|^ 2d\rho d\theta, \] so that A(r,f) is the area of f(\(| z| \leq r)\) taking account of multiplicity. For functions \(f\in H^ 2\) the authors note that \((1)\quad (1-r)A(r,f)=o(1)\) (r\(\to 1)\); and, if \(\phi\) is a positive function on (0,1) satisfying (2) \(\int^{1}_{0}\phi (t)dt=\infty\), then \((3)\quad A(r,f)=o(\phi (r))\) (r\(\to 1\), \(r\in S_{\phi})\), where \(S_{\phi}\) is a subset of (0,1) containing points arbitrarily near 1; and \((4)\quad \int_{E_{\phi}}\phi (t)dt<\infty\) where \(E_{\phi}\) is the set of r in (0,1) for which A(r,f)\(\geq \phi (r).\) The object of the paper is to show the extent to which these inequalities are best possible for bounded functions f. It is shown that the o(1) function in (1) cannot be made specific; that for smoothly behaved functions \(\phi\) satisfying (2) the estimate for \(E_{\phi}\) in (4) is sharp in a natural sense; that (3) need not be true if (2) is relaxed; that any upper bound applicable to \(H^ 2\) functions cannot be improved for bounded functions; and that the counterexamples constructed cannot in general have positive coefficients.











This page was built for publication: On the Area of the Image of a Bounded Analytic Function

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3681097)