On successive coefficients of univalent functions

From MaRDI portal





Let \(f(z)=z+\sum^ \infty_{n=2}a_ nz^ n\) be univalent in the unit disk, and set \[ h(z)=(f(z^ 2))^{1/2}=\sum^ \infty_{n=0}c_{2n+1}z^{2n+1}. \] A well known conjecture asserts that \(d(n)\equiv| c_{2n+1}|-| c_{2n-1}|=O(n^{- 1/2})\) as \(n\to\infty\). The functions \((k(z^ 4))^{1/4}\), where \(k\) denotes the Koebe function, show that the decay rate \(n^{-1/2}\) cannot be improved. The best known order of decay, due to K. W. Lucas (1969), is \(O(n^{1-\sqrt 2})\). The author proves some results related to the conjecture provided the function \(f\) grows either rapidly enough or slowly enough. Theorem 2 asserts that if \(f(z)=O(1-r)^{-1}\) as \(r\to 1\) then \(d(n)=O(n^{- 1/2})\), and if \(\lim\inf_{r\to 1}(1-r)| f(re^{i\theta_ 0})|>0\) for some \(\theta_ 0\) then \(d_ n=O(n^{-1/2}\log n)\). In Theorem 1, he proves that if \(\lim\inf_{r\to 1}(1-r)^ 2\max_ \theta| f(re^{i\theta})|>0\) then \(d(n)=O(n^{-1/2- \varepsilon})\), where \(\varepsilon=1/320\). It had been known previously that for such functions one does have \(d(n)=O(n^{-1/2}\)).











This page was built for publication: On successive coefficients of univalent functions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3991040)