Range-kernel characterizations of operators which are adjoint of each other (Q782537): Difference between revisions

Let \(H\) and \(K\) be two Hilbert spaces and let \(S:H\to K\) and \(T:K\to H\) be (not necessarily densely defined or closed) linear operators such that \[\operatorname{ran} (S\cap T^*)=KX\text{ and }\operatorname{ran}(T\cap S^*)=H.\] The authors prove that \(S\) and \(T\) are both densely defined operators such that \(S^*=T\) and \(T^*=S\). They also show that, if \(S\) is a densely defined operator, then the following statements are equivalent: (i) \(S\) is closed, (ii) \(S^*S\) and \(SS^*\) are self-adjoint operators, (iii) \(\operatorname{ran} (I+S^*S)=H\) and \(\operatorname{ran} (I+SS^*)=H\). The authors also generalize some analogous results from \textit{T. Nieminen} [Ann. Acad. Sci. Fenn., Ser. A I 316, 5 p. (1962; Zbl 0171.34502)] and \textit{M. Roman} and \textit{A. Sandovici} [Result. Math. 74, No. 2, Paper No. 73, 6 p. (2019; Zbl 1504.47010)] and give some characterizations of self-adjoint, skew-adjoint and unitary operators.