Factorization of the Fock–Sobolev space (Q6584641)

Let \(0<p\le \infty\) and let \(F^p\) denote the \emph{Fock space} consisting of all entire functions \(f\) such that \N\[\N\Vert f\Vert^p_{F^p}=\frac{p}{2}\int_{\mathbb{C}}\left |f(z)e^{-\frac{1}{2}|z|^2}\right |^p dA(z)<\infty\N\]\Nwhere \(dA(z)=\pi^{-1}dxdy\). For \(p=\infty\), the norm in \(F^\infty\) is defined by \N\[\N\Vert f\Vert_{F^\infty}=\sup_{z\in\mathbb{C}}|f(z)|e^{-\frac{1}{2}|z|^2}.\N\]\NFor a positive integer \(m\), the \textit{Fock-Sobolev space} \(F^{p,m}\) consists of all entire functions \(f\) such that \(f^{(m)}\), the \(m\)th order derivative of \(f\), belongs to the Fock space \(F^p\). It is known (see [\textit{H. R. Cho} et al., Potential Anal. 43, No. 2, 199--240 (2015; Zbl 1327.32009); \textit{H. R. Cho} and \textit{S. Park}, Nagoya Math. J. 237, 79--97 (2020; Zbl 1442.32012); \textit{H. R. Cho} and \textit{K. Zhu}, J. Funct. Anal. 263, No. 8, 2483--2506 (2012; Zbl 1264.46017)]) that \N\[\Nf\in F^{p,m} \iff z^mf\in F^p.\N\]\NThe \textit{weighted Fock-Sobolev space} \(F^p_m\) consists of all entire functions \(f\) that satisfy \N\[\N\int_{\mathbb{C}}\left |(1+|z|)^mf(z)e^{-\frac{1}{2}|z|^2}\right |^p dA(z)<\infty\N\]\Nfor \(0<p<\infty\), and \N\[\N\sup_{z\in\mathbb{C}}(1+|z|)^m|f(z)|e^{-\frac{1}{2}|z|^2}<\infty\N\]\Nfor \(p=\infty\). It is known that \(F^p_m=F^{p,m}\), and \N\[\NF^p_m\subset F^p\subset F^p_{-m}.\N\]\NIn the paper under review, the authors prove the following factorization theorem for weighted Fock spaces with polynomially growing/decaying weights: \N\NTheorem. For \(0<p\le \infty\), and \(m\in\mathbb{N}\) we have \[F^p=\mathcal{P}_m\cdot F^p_m,\qquad F^p_{-m}=\mathcal{P}_m\cdot F^p,\] where \(\mathcal{P}_m\) is the set of polynomials having degrees less than or equal to \(m\).\NThe proof is based on some pointwise estimates and the Hadamard factorization theorem for entire functions.