Point localization and spectral theory for symmetric random operators (Q1024178): Difference between revisions

Let \(H\) be a Hilbert space, \((\Omega ,\mathfrak{A} ,\mu )\) be a probability space, and \(M(H)\) the space of equivalent classes of functions \(f:\Omega \to H\) endowed with the convergence in measure. By a random operator, the author means a densely defined linear operator \(A:D_A\to M(H)\), where \(D_A\subset H\). Since such an \(A\) is not an endomorphism, the notions of spectral theory do not immediately apply. The random operator \(A\) can be represented as a measurable field of closed operators \((B(\omega ))_{\omega \in \Omega}\) in \(H\). The author proposes to treat the study of probabilistic characteristics of the spectrum of \(B(\omega )\) as a ``spectral theory'' for \(A\).