The Radon-Nikodym theorem for random measures (Q1824935)

A random measure \(\mu\) is said to be absolutely continuous with respect to a random measure \(\nu\) if they are defined on the same spaces (X,\({\mathcal B})\) and (\(\Omega\),F,P) and if there exists a function \(f(x,\omega):\quad X\times \Omega \to R\), which, for each fixed \(\omega\), is \({\mathcal B}\)-measurable and is such that \[ (1)\quad \forall B\in {\mathcal B}:\quad \mu (B,\omega)=\int_{B}f(x,\omega)\nu (dx,w)\quad a.s. \] If \(\mu\) is absolutely continuous with respect to \(\nu\), then \[ (2)\quad \forall B\in {\mathcal B}:\quad \nu (B,\omega)=0\quad a.s.\quad \Rightarrow \quad \mu (B,\omega)=0\quad a.s. \] At the same time, (2) is not a sufficient condition for absolute continuity. Let \(\mu (\cdot,\omega_ 0)<<\nu (\cdot,\omega_ 0)\) denote the ordinary absolute continuity of the measure \(\mu (\cdot,\omega_ 0)\) with respect to the measure \(\nu (\cdot,\omega_ 0)\). One result of the paper is the Theorem: A random measure \(\mu\) is absolutely continuous with respect to a random measure \(\nu\) if and only if \(\mu (\cdot,\omega_ 0)<<\nu (\cdot,\omega_ 0)\) for almost each fixed \(\omega_ 0\in \Omega\).