Statistical extensions of some classical Tauberian theorems for Cesàro summability of triple sequences (Q346584)

From MaRDI portal
Revision as of 14:56, 9 December 2024 by Import241208061232 (talk | contribs) (Normalize DOI.)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
scientific article
Language Label Description Also known as
English
Statistical extensions of some classical Tauberian theorems for Cesàro summability of triple sequences
scientific article

    Statements

    Statistical extensions of some classical Tauberian theorems for Cesàro summability of triple sequences (English)
    0 references
    0 references
    0 references
    0 references
    29 November 2016
    0 references
    A sequence \(u=(u_{m})\) is called \((C,1)\) summable to a finite real number \(L\) if the sequence \(\sigma_{m}(u)\) is convergent to \(L\), where the \((C,1)\) means of a sequence \((u_{m})\) is \(\sigma_{m}(u)=\frac{1}{m+1}\sum_{i=0}^{m}u_{i}\). A triple sequence \((u_{mns})\) of real numbers is said to be convergent in Pringsheim's sense (or \(P\)-convergent) to a finite real number \(L\) if, for a given \(\epsilon > 0 \), there exists an integer \(N> 0\) such that \(|u_{mns} - L| < \epsilon \) whenever \(m, n,s \geq N\). The \((C,1,1,1)\), \((C,1,0,0)\), \((C,0,1,0)\) and \((C,0,0,1)\) means of a sequence \((u_{mns})\) are defined, respectively, by \[ \sigma_{mns}^{(111)}(u)=\frac{1}{(m+1)(n+1)(s+1)}\sum_{i=0}^{m}\sum_{j=0}^{n}\sum_{k=0}^{s}u_{ijk}, \] \[ \sigma_{mns}^{(100)}(u)=\frac{1}{m+1}\sum_{i=0}^{m}u_{ins}, \] \[ \sigma_{mns}^{(010)}(u)=\frac{1}{n+1}\sum_{j=0}^{n}u_{mjs}, \] \[ \sigma_{mns}^{(001)}(u)=\frac{1}{s+1}\sum_{k=0}^{s}u_{mnk}, \] for all nonnegative integers \(m, n,s\). A sequence \((u_{mns})\) is called \((C,1,1,1)\) summable to a finite real number \(L\) if the sequence \((\sigma_{mns}^{(111)}(u))\) is \(P\)-convergent to \(L\). \((C,1,0,0)\), \((C,0,1,0)\) and \((C,0,0,1)\) summability of \((u_{mns})\) are defined similarly as the \(P\)-convergence of the sequences \((\sigma_{mns}^{(100)}(u))\), \((\sigma_{mns}^{(010)}(u))\), and \((\sigma_{mns}^{(001)}(u))\) to \(L\), respectively. A triple sequence \((u_{mns})\) is bounded if there exists \(M > 0\) such that \(|u_{mns}|\leq M\) for all nonnegative integers \(m,n,s\). A sequence \((u_{mns})\) is said to be statistically convergent to \(L\) if, for arbitrary \(\epsilon>0\), \[ \lim_{m,n,s\to \infty}\frac{1}{(m+1)(n+1)(s+1)}\left| \{(i, j, k):|u_{ijk} -L|\geq \epsilon, \, i\leq m,\, j\leq n,\, k\leq s \}\right|=0. \] On the other hand, a sequence \((u_{mns})\) is said to be slowly oscillating and statistically slowly oscillating in sense \((1,0,0)\) if \[ \lim_{\lambda\to 1^{+}}\limsup_{m,n,s\to \infty}\max_{m+1\leq i\leq \lambda_{m}}\left|\sum_{r=m+1}^{i}\Delta_{r}u_{r,n,s}\right|=0, \] and for every \(\epsilon>0\), \[ \begin{multlined} \inf_{\lambda>1}\limsup_{M,N,S\to \infty}\frac{1}{(M+1)(N+1)(S+1)}|\{(m, n, s) :\\ \max_{m+1\leq i\leq \lambda_{m}} |u_{ins}-u_{mns}| \geq \epsilon, m\leq M, n\leq N, s\leq S\}|=0, \end{multlined} \] where \(\lambda_{m}\) indicates the integer part of \(\lambda m\), respectively. The concepts of slow oscillation and statistical slow oscillation for \((u_{mns})\) in senses \((0,1,0)\) and \((0,0,1)\) are defined analogously. The results in this paper are essentially concerned with the question under which conditions statistically convergent and statistically \((C,1,1,1)\) summable triple sequences are \(P\)-convergent. In this regard, the authors indicate that the statistical convergence implies \(P\)-convergence for triple sequences under the two-sided boundedness conditions and slowly oscillating conditions in certain senses. Moreover, they reveal the relation between statistically \((C,1,1,1)\) summable triple sequences and \(P\)-convergent triple sequences. In addition to these, the authors clearly state that bounded and statistically \((C,1,1,1)\) summable triple sequences are \(P\)-convergent under the two-sided boundedness conditions and slowly oscillating conditions in certain senses. Replacing the conditions of slow oscillation by the conditions of statistical slow oscillation in certain senses, the authors obtain the statistical convergence of a sequence from its statistical \((C,1,1,1)\) summability without the boundedness condition on the sequence.
    0 references
    triple sequences
    0 references
    Tauberian theorems
    0 references
    slowly oscillating sequences
    0 references
    statistical convergence
    0 references
    convergence in Pringsheim's sense
    0 references

    Identifiers