Some sequence spaces and statistical convergence (Q1607849)

Let \(\lambda= (\lambda_n)\) be a real sequence with \(1= \lambda_1\leq \lambda_2\leq\cdots\), \(\lim_n \lambda_n=+\infty\), \(\lambda_{n+1}\leq \lambda_n+ 1\) \((n= 1,2,\dots)\). Using the mean of de la Vallée Poussin, the author defines the concept of strong \((V,\lambda)\)-summability. A sequence \(x= (x_n)\) is said to be strongly \((V,\lambda)\)-summable if there exists an \(L\) such that \[ t_n(x)= {1\over \lambda_n} \sum_{k\in I_n}|x_k- L|\to 0\qquad (n\to\infty), \] where \(I_n= [n- \lambda_n+ 1,n]\) \((n= 1,2,\dots)\). Denote by \([V,\lambda]\) the set of all strongly \((V,\lambda)\)-summable sequences. This concept can be extended to the concept of strongly \((V,\lambda)\)-summable sequences with respect to a modulus \(f\) (\(f:[0,+\infty)\to [0,+\infty)\), \(f(0)= 0\), \(f\) increasing and continuous at \(0\) from the right). We get this extension if we replace \(|x_k- L|\) by \(f(|x_k- L|)\). So we get the set \([V,\lambda, f]\) of all strongly \(\lambda\)-summable sequences with respect to \(f\). A further extension can be obtained by using the concept of statistical convergence; cf. \textit{H. Fast} [Colloq. Math. 2, 241-244 (1951; Zbl 0044.33605)]. A sequence \(x= (x_k)\) is said to be \(\lambda\)-statistically convergent if there is an \(L\) such that for every \(\varepsilon> 0\) we have \[ \lim_n {1\over\lambda_n} |\{k\in I_n:|x_k- L|\geq \varepsilon\}|= 0, \] where \(|M|\) is the cardinality of \(M\). Denote by \(s_\lambda\) the set of all \(\lambda\)-statistically convergent sequences. The author shows that \([V, \lambda, f]= [V,\lambda]\) provided that \(\lim_{t\to\infty} {f(t)\over t}> 0\) and \([V,\lambda,f]= s_\lambda\) if and only if \(f\) is a bounded function.