On the growth of zero-free MacLane-universal entire functions (Q441349): Difference between revisions

The authors consider the growth of zero-free entire functions which are universal in the sense of MacLane. Let \(X\), \(Y\) be two topological spaces, \(T_n: X\to Y\), \(n\in \mathbb{N}\), a sequence of continuous mappings, and \(x_0\in X\). Then \(x_0\) is said to be universal for \((T_n)\) if its orbit \(\{T_n x_0: n\in \mathbb{N}\}\) under \((T_n)\) is dense in \(Y\). If \(X=Y\) and \(X\) is a topological vector space, then an operator \(T\) on \(X\) is said to be hypercyclic provided that the sequence \((T_n)\) of iterates is universal. Consider the derivative operator \(D:f\in H(\mathbb C)\to H(\mathbb C)\), where \(H(\mathbb C)\) is a set of entire functions. \textit{G. R. MacLane} showed that \(D\) is hypercyclic~[J. Anal. Math. 2, 72--87 (1952; Zbl 0049.05603)]. For a function \(\varphi: (0,+\infty)\to(0,+\infty)\) with \(\lim_{r\to\infty}\varphi(r)=\infty\), the exponential growth order and the lower exponential growth order are respectively defined by \(\rho(\varphi)=\limsup_{r\to\infty}\frac{\log\log\varphi(r)}{\log r}\) and \(\tilde\rho(\varphi)=\liminf_{r\to\infty}\frac{\log\log\varphi(r)}{\log r}\). The main results in this paper are the following: (a) There is no \(D\)-hypercyclic entire function \(f\) with finitely many zeros satisfying \(\tilde\rho(f)<+\infty\). (b) If \(\varphi: (0,+\infty)\to(0,+\infty)\) is a function with \(\rho(\varphi)=+\infty\), then there is a \(D\)-hypercyclic zero free entire function \(f\) such that \(\liminf_{r\to\infty}\frac{M(r,f)}{\varphi(r)}<+\infty\).