On the numerical range of a Volterra operator in \(L_p\) (Q5926411)

From MaRDI portal
scientific article; zbMATH DE number 1571139
Language Label Description Also known as
English
On the numerical range of a Volterra operator in \(L_p\)
scientific article; zbMATH DE number 1571139

    Statements

    On the numerical range of a Volterra operator in \(L_p\) (English)
    0 references
    0 references
    0 references
    17 December 2001
    0 references
    Let \(A\) be a bounded linear operator on the complex Lebesgue space \(L_p=L_p(\Omega, \Xi, \mu)\). The numerical range \(A\) is defined as \(W(A)=\{ \langle A(x), J_p(x)\rangle: \|x\|_p=1\}\) where the duality map \(J_p\) for \(L_p\) into its dual \(L_p^{\ast}\) is given by \(J_p(x)=\|x\|^{2-p}_p\cdot |x|^{p-1}\cdot \text{sgn } \overline{x}\). Earlier in [Integral Equations Oper. Theory 31, No. 4, 421-435 (1998; Zbl 0917.47004)] the author, \textit{F. Okoh} and \textit{G. K. R. Rao} indicated some examples of operators \(A\) in \(L_p(\Omega, \Xi, \mu)\) for which the supremum \(\Lambda(A)=\sup\{ \Re\;\alpha: \alpha \in W(A)\}\) is attained as a maximum in the cases where the underlying measure \(\mu\) was discrete. The basic subject of this paper is deriving an example \(A\) in the space \(L_p(0, 1)\). Let \(A\) be the Volterra-type operator defined by \((Ax)(s)=\int_0^s x(t) dt\), \(x\in L_p(0, 1)\), \(1<p<\infty\), \(s\in [0, 1]\). The author shows that \(\Lambda(A)\) have maximum \((1+{2\log (p/q)\over{p-q}})^{-1}\) , where \(p^{-1}+q^{-1}=1\). The set of all maximizing functions is also specified.
    0 references
    0 references
    Lebesgue space
    0 references
    Volterra operator
    0 references
    numerical range
    0 references
    eigenvalue problem
    0 references