Growth of entire functions of exponential type defined by convolution (Q2431752)

If \(f(z) = \sum_{n=0}^\infty a_n z^n\) is an entire function of exponential type, then its Borel transform \(B(f)\) is defined by \(B(f)(z) = \sum_{n=0}^\infty n!a_n z^{-n-1}\) (\(B(f)\) converges on \(| z| > \tau\) if and only if \(f\) is of type \(\tau\)), and its indicator function \(h_f\) is defined as \(h_f(z) = \limsup_{r \to \infty} {\log | f(rz)| \over r}\) \((z \in\mathbb C\setminus \{0\})\). For a compact convex set \(L \subset\mathbb C\), the function \(H_L\) defined for \(z \in\mathbb C\) by \(H_L(z) = \sup_{s \in L}\operatorname{Re} (zs)\) is called the support function of \(L\). For \(f \neq 0\), \(h_f\) is the support function of a compact convex set \(I_f\), which coincides with the smallest closed convex set outside which \(B(f)\) is a holomorphic function. Given \(L\) as before, by Exp\((L)\) it is denoted the vector space of all entire functions \(f\) such that for every \(\varepsilon > 0\) there is a constant \(C_\varepsilon\) such that \(| f(z)| \leq C_\varepsilon \exp (H_L(z) + \varepsilon | z| )\) \((z \in\mathbb C)\). Then \(f \in \text{Exp}(L)\) if and only if \(h_f(z) \leq H_L(z)\), and if and only if \(I_f \subset L\). For two functions \(f\) holomorphic in \(| z| <r\) and \(g\) holomorphic in \(| z| <R\), with respective Taylor coefficients \(a_n, b_n\) at the origin, the Hadamard convolution \(f*g\) of \(f\) and \(g\) is defined by \((f*g)(z) = \sum_{n=0}^\infty a_n b_n z^n\). Let \(A \subset\mathbb C\). By \(A^{-1}\), \(cv(A)\) it is denoted, respectively, the set \(\{z^{-1}: z \in A\}\) and the convex hull of \(A\), while for an open set \(\Omega \subset\mathbb C_\infty\) \((:=\mathbb C\cup \{\infty\})\), \(H_0(\Omega )\) stands for the set of all functions holomorphic on \(\Omega\) and vanishing at \(\infty\) if \(\infty \in \Omega\). For \(\Omega \subset\mathbb C\), \(H(\Omega )\) is the space of holomorphic functions on \(\Omega\). Based on the Borel transformation and the Hadamard multiplication theorem on singularities of the Hadamard convolution, the authors obtain a number of results on the growth of entire functions defined by convolution of an entire function of exponential type with a function holomorphic at the origin. The following statements are among their results: (a) Let \(L \subset\mathbb C\) be a convex compact set and \(\phi \in H_0(\mathbb C\setminus K^{-1})\), where \(K \subset\mathbb C\) is a compact set. If \(f \in \text{Exp}(L)\), then \(f * \phi \in \text{Exp}(cv(L \cdot K))\). In particular, \(h_{f*\phi} \leq \max \{0,h_f\}\) (\(\leq h_f\)) whenever \(\phi \in H(\mathbb C\setminus [1,\infty ))\) (whenever \(\phi \in H_0(\mathbb C\setminus \{1\}\), resp.) (b) Let \(\phi \in H_0(\mathbb C\setminus \{1\})\) with Taylor coefficients \(\phi_n \neq 0\) at the origin, and let \(f\) be an entire function of exponential type. Then \(h_{f*\phi} = h_f\) if some of the following two conditions is satisfied. 1. Zero belongs to \(I_f\) but is not an extreme point of \(I_f\), and \(M_\alpha := \{z=re^{i\theta}: r \geq 0\), \(| \theta | \leq \alpha\} \cap \{\)zeros of the Mellin transform of \(f\}\) is finite for all \(\alpha < \pi /2\). 2. Zero does not belong to \(I_f\) and \(M_\alpha\) is finite for all \(\alpha < \pi\). More specialized statements on convolution with special (mainly, hypergeometric) functions or with entire functions of more general order are also obtained.