More about the kernel convergence and the ideal convergence (Q383663): Difference between revisions

Let \(\mathcal{I}\) be a proper ideal of subsets of \(\mathbb{N}\) and let \(p:\ell_{\infty}\rightarrow\mathbb{R}\) a seminorm with \(p\left( \chi_{\mathbb{N}}\right) =\sup\left\{ p\left( t\right) :\left\| t\right\| _{\infty}\leq1\right\} \). A sequence \(\left( x_{n}\right) \) of points of a real Banach space \(X\) is said to be\newline1) \(\mathcal{I} \)-convergent to \(x\in X\) if for each \(\varepsilon>0\) the set \(A(\varepsilon ):=\left\{ n\in \mathbb{N}:\left| x_{n}-x\right| \geq\varepsilon\right\} \) belongs to \(\mathcal{I}\), and \newline2) \(p\)-kernel convergent to \(x\in X\) if \(p\left( \chi_{A\left( \varepsilon\right) }\right) =0\) for each \(\varepsilon>0.\) \newline The \(\mathcal{I}_{d}\)-convergence with \(\mathcal{I}_{d}:=\left\{ A\subset\mathbb{N}:\lim_{n\rightarrow\infty} \frac{\left| \{ k\in A \text{ } : \text{ } k\leq n \}\right| }{n}=0\right\} \) is known as statistical convergence. The authors show that the \(\mathcal{I} \)-convergence is equivalent to the \(p_{\mathcal{I}}\)-kernel convergence, where \(p_{\mathcal{I}}\) is the quotient norm of \(\ell_{\infty}/\overline {\text{span}}\left\{ \chi_{A}:A\in\mathcal{I}\right\} \) (Theorem 3.7). Let \(p\left( t\right) :=\limsup_{n\rightarrow\infty}\frac{1}{n}\sum _{k=1}^{n}\left| t_{k}\right| \) \(\left( t=\left( t_{k}\right) \in\ell_{\infty}\right) \) and \(\mu:=t^{\ast}\circ\chi_{\left( \cdot\right) }\), where \(t^{\ast}\in\left( \ell_{\infty}\right) ^{\ast}\) is such that \(t^{\ast}\leq p\) and \(\left\langle \chi_{\mathbb{N}},t^{\ast}\right\rangle =1.\) Then \(\mathcal{I}_{\mu}:=\left\{ A\subset\mathbb{N}:\mu\left( A\right) =0\right\} \) is a proper ideal and for a sequence \(\left( x_{n}\right) \) in a Banach space \(X\), the following three statements are equivalent: \(\left( 1\right) \) \(\left( x_{n}\right) \) is statistically convergent to \(x,\) \(\left( 2\right) \) \(\left( x_{n}\right) \) is \(p\)-kernel convergent to \(x,\) \(\left( 3\right) \) \(\left( x_{n}\right) \) is \(\mathcal{I}_{\mu} \)-convergent to \(x\) (Theorem 4.3).