Highly tempering infinite matrices. II: From divergence to convergence via Toeplitz-Silverman matrices (Q2204991)

A Toeplitz-Silverman (T-S) matrix is an infinite matrix \(A\) with rows and columns indexed by the natural numbers \(\mathbb{N}\) such that: (a) the series defined by each row is absolutely convergent with sum \(1\); and (b) the sequence defined by each column converges to \(0\). It is known that a matrix \(A\) is T-S if and only if whenever the sequence of entries of a column vector \(\mathbf{x}\) converges then \(A\mathbf{x}\) converges to the same limit (see [\textit{A. Peyerimhoff}, Lectures on summability. Berlin-Heidelberg-New York: Springer Verlag (1969; Zbl 0182.08401); \textit{A. Wilansky}, Summability through functional analysis. Amsterdam-New York-Oxford: North-Holland (1984; Zbl 0531.40008)]). For each real T-S matrix \(A\), let \(\Omega_{A}\) denote the set of all \(\mathbf{x}\in\mathbb{R}^{\mathbb{N}}\) for which \(A\mathbf{x}\) converges but \(\mathbf{x}\) does not. The authors show that \(\Omega_{A}\) is large in a suitable sense, which is too technical to be stated here. For example Corollary 3.3 states: \(\Omega_{A}\) is \(\mathbf{c}\)-algebrable, \(\aleph_{0}\)-latticeable and dense in \(\mathbb{R}^{\mathbb{N}}\). The paper concludes by considering a similar problem in the ultrametric setting. For Part I see [\textit{L. Bernal-González} et al., ibid. 112, No. 2, 341--345 (2018; Zbl 1390.15093)].