On the absolute Hardy-Bohr criteria (Q1288316)

For an FK space \(E\) containing \(\varphi\) (the finitely nonzero sequences) the \(f\)-dual is defined as \(E^f:= \{(f(e_i))\mid f\in E'\}\), where \(E'\) is the topological dual of \(E\) and \(e_i\) denotes the \(i\)th coordinate sequence. \(E\) is called a sum space [see \textit{W. H. Ruckle}, Can. J. Math. 22, 863-874 (1970; Zbl 0201.15702)] if \(E^f\) coincides with sequence space \(\{y= (y_k)\mid (x_ky_k)\in E\) for all \(x\in E\}\) (the multipliers on \(E)\). A study of sum spaces in the context of summability is motivated by the classical problem of summability factors. For a regular normal matrix method \(A=(a_{nk}) \) the series to sequence convergence domain \(c_{A\Sigma}\) (here \(\Sigma\) is the triangle of ones) is a sum space if and only if the Hardy-Bohr criterion determines the summability factors of \(A\). Similarly, if \(A\) is absolutely regular (i.e. regular and \(A:bv\to bv\) (the absolutely convergent sequences)), the absolute summability factors are characterized by the Hardy-Bohr conditions if and only if the variational summability domain \(bv_{A \Sigma}\) is a sum space [cf. the authors, Math. Z. 199, No. 1, 99-108 (1988; Zbl 0631.40008), the second author, Math. Z. 194, 405-414 (1987; Zbl 0596.40006)]. The authors show that in this situation the method \(A\) is of type \(M(bv_0)\) (i.e. if \(\sum_n t_na_{nk}= 0\) \((k=0,1, \dots)\) for \(t=(t_n)\in bs\) (the bounded series), then \(t=0)\) (Proposition 4.1). Using the notion of a \(T\)-solid sequence space [compare \textit{K.-G. Grosse-Erdmann}, Result. Math. 23, No. 3-4, 303-321 (1993; Zbl 0806.46009)] they prove that for a regular Nörlund polynomial method \(N_p\) with \(bv_{N_p}\neq bv\) the BK space \(bv_{N_p\Sigma}\) is not a sum space (Theorem 3.1). The Nörlund method \(N_p\) defined by \(p(z)= (1+z)/(1-z)^r\) \((r\) a positive integer) is of type \(M(bv_0)\) (Proposition 5.1). An open question is whether the variational summability domain \(bv_{N_p\Sigma}\) for this method is a sum space.