Lower bounds for summability matrices on weighted sequence spaces (Q2470029): Difference between revisions

Let \(A\) be the linear operator defined by a nonnegative summability matrix \(A=(a_{nk}) \) with (i) \(a_{nk}=0\) for \(k>n\) and (ii) \(\sum_{k=1}^{n}a_{nk}=1\) for each \(n\in \mathbb{N}\). The authors consider the operators \(A\) on the sequence spaces \(\ell _{p}(w) =\{ x=(x_{n}) \mid \| x\| _{p,w}=(\sum_{n}w_{n}| x_{n}| ^{p}) ^{1/p}<\infty \} \) (here, \((w_{n}) \) is a decreasing nonnegative sequence and \(1\leq p<\infty),\) on the Lorentz sequence spaces \( d(w,1) =\{ (x_{n}) \mid \| x\| _{1,w}=\sum_{n}w_{n}x_{n}^{\ast }<\infty \} \) (here, \((x_{n}^{\ast }) \) is a decreasing rearrangement of \((| x_{n}|) \), \(w_{n}\to 0\) and \(\sum_{n}w_{n}=\infty)\), and on the conjugate space of \(d(w,1) \) (denoted by \(e(w,\infty)),\) and seek largest possible lower bounds \(L(A) >0\) such that \(\| Ax\| \geq L(A) \| x\| \) holds for all decreasing nonnegative sequences \( x=(x_{n}) .\) For example, it is proved that for the operators \( A:e(w,\infty) \to e(w,\infty) \) we have \( (a) \) \(L(A^{t}) =1,\) where \(A^{t}\) is the transpose matrix of \(A,\) and \((b) \) \(L(A) =1,\) if \(w_{1}=1\) and \(w_{n}\geq a_{n1}\) for all \(n\in \mathbb{N}\).