Matrix maps of quasi-convex null sequences (Q1848166)

Let \((E,p_{\mu(\mu\in\mathbb{N})})\) be an arbitrary FK-space and let \(L=(e_{nk})\) be the matrix defined by \[ e_{nk}=\begin{cases} E^{\alpha}_{k-1-n}/E^{\alpha}_{k-1} & \text{if} 0 \leq n < k, \\ 0 & \text{otherwise}, \end{cases} \] where \[ E^{\alpha}_{m} = \begin{pmatrix} m + \alpha - 1 \\ m \end{pmatrix}, \alpha\in\mathbb{N}, m\in\mathbb{N}\cup \{0\}=\mathbb{N}_{0} \] Let us consider \(q^{\alpha}\) -- the space of all quasi -- convex (of order \(x\)) convergent sequences, \(q^{\alpha}_{0}\) -- the space of all quasi - convex null sequences and \(m_{S}\) -- the space of all sequences \(x=(x_{n})\) such that \[ \|x\|_{m_{S}}=\sup_{m}\bigg|\sum^{m}_{n=0}x_{n}\bigg|< \infty. \] The author establishes the following main result: \textbf{Theorem 2.3.} A matrix \(A=(a_{nk})\) maps \(q^{\alpha}_{0}\) into \(E\) if and only if the following conditions hold: (i) \[ (a_{nk})_{k} \in m_{S} \text{ for each } n \in \mathbb{N}_{0}; \] (ii) \[ (a_{nk})_{n} \in E \text{ for each } k \in \mathbb{N}_{0}; \] (iii) \[ \sup_{k}p_{\mu} \big( \big(\sum^{k}_{\nu=0}a_{n\nu}e_{\nu k+1} \big)_n \big) < \infty \hbox{ for each } \mu\in \mathbb{N}. \] If \(A\in (q^{\alpha}_{0},E)\) and \(E \subset c\), then \(a_{k}=\lim_{n}a_{nk}\) exists for each \(k\in \mathbb{N}_{0}\) and we get \[ \lim A_{x}= \sum_{n}\Delta^{\alpha}x_{n}\sum^{n}_{k=0}E^{\alpha}_{n-k}a_{k} \text{ for every } \;x \in q^{\alpha}_{0}. \] Consequently, the author obtains the characterization of matrices \(A\in (q^{\alpha}_{0},E)\), where \(E\) is one of the FK - spaces considered by \textit{M. Stieglitz} and \textit{H. Tietz} [Math. Z. 154, 1-16 (1977; Zbl 0345.40004)]. Moreover, the author determines the matrices which map these FK - spaces into \(q^{\alpha}_{0}\). Finally, one remarks that some conditions in the paper by Stieglitz and Tietz may be omitted.