Table of Contents
Fetching ...

Splitting of Liftings in Product Spaces II

Kazimierz Musial

TL;DR

Let $R\in P\circledast Q$ be a skew product of two probability spaces with a $Q$-disintegration $\{({\mathfrak A}_y,S_y):y\in Y\}$. The paper constructs liftings $\pi$ on ${\mathfrak A}\divideontimes{\mathfrak B}$ and fiberwise liftings $\sigma_y$ on $\widehat{S_y}$ so that $[\pi(f)]^y=\sigma_y([\pi(f)]^y)$ for all $f\in{\mathcal L}^{\infty}(\widehat{R_{\divideontimes}})$ and a.e. $y$; it proves the main theorems (notably mu25) to connect conditional expectations across the fibers with the liftings. Under absolute continuity assumptions $R\ll P\times Q$ (or $S_y\ll P|_{\mathfrak A_y}$), the liftings can be taken to take values in $\mathfrak A\widehat{\otimes}\mathfrak B$, and the work both generalizes and corrects earlier results such as Theorem 3.8 in mu25. Finally, the paper applies the framework to characterize stochastic processes possessing equivalent measurable versions and to derive measurability-preserving modifications.

Abstract

Let $(X, \mfA,P)$ and $(Y, \mfB,Q)$ be two probability spaces, $R$ be their skew product on the product $σ$-algebra $\mfA\otimes\mfB$ and $\{(\mfA_y,S_y)\colon y\in{Y}\}$ be a $Q$-disintegration of $R$. Then let $\mfA\dd\mfB$ be the $σ$-algebra generated $\mfA\otimes\mfB$ and by the family $\mcM:=\{E\subset{X\times{Y}}\colon \exists\;N\in\mfB_0\;\forall\;y\notin{N}\;\wh{S_y}(E^y)=0\}$ and $\wh{R_{\dd}}$ be the extension of $R$ such that $\mcM$ becomes the family of $\wh{R_*}$-zero sets ($\wh{S_y}$ is the completion of $S_y$ and $\mfB_0=\{B\in\mfB: Q(B)=0\}$). We prove that there exist a lifting $π$ on $\mcL^{\infty}(\wh{R_{\dd}})$ and liftings $σ_y$ on $\mcL^{\infty}(\wh{S_y})$ , $y\in Y$, such that \[ [π(f)]^y= σ_y\Bigl([π(f)]^y\Bigr) \qquad\mbox{for every} \quad y\in Y\quad\mbox{and every}\quad f\in\mcL^{\infty}(\wh{R_{\dd}}). \] In case of a separable $P$ and in case when $R\ll{P}\times{Q}$ a characterization of stochastic processes possessing an equivalent measurable version is presented. The theorem is a generalization and correction of \cite[Theorem 3.8]{mu25}.

Splitting of Liftings in Product Spaces II

TL;DR

Let be a skew product of two probability spaces with a -disintegration . The paper constructs liftings on and fiberwise liftings on so that for all and a.e. ; it proves the main theorems (notably mu25) to connect conditional expectations across the fibers with the liftings. Under absolute continuity assumptions (or ), the liftings can be taken to take values in , and the work both generalizes and corrects earlier results such as Theorem 3.8 in mu25. Finally, the paper applies the framework to characterize stochastic processes possessing equivalent measurable versions and to derive measurability-preserving modifications.

Abstract

Let and be two probability spaces, be their skew product on the product -algebra and be a -disintegration of . Then let be the -algebra generated and by the family and be the extension of such that becomes the family of -zero sets ( is the completion of and ). We prove that there exist a lifting on and liftings on , , such that \[ [π(f)]^y= σ_y\Bigl([π(f)]^y\Bigr) \qquad\mbox{for every} \quad y\in Y\quad\mbox{and every}\quad f\in\mcL^{\infty}(\wh{R_{\dd}}). \] In case of a separable and in case when a characterization of stochastic processes possessing an equivalent measurable version is presented. The theorem is a generalization and correction of \cite[Theorem 3.8]{mu25}.
Paper Structure (5 sections, 12 theorems, 36 equations)

This paper contains 5 sections, 12 theorems, 36 equations.

Key Result

Proposition 1.2

Pachl1 Let $(X,{\mathfrak A},P)$ be a probability space. $P$ is compact if and only if for every complete probability space $(Y,{\mathfrak B},Q)$ and every $R\in{P}\circledast{Q}$ there exists a $Q$-disintegration of $R$. If $P$ is compact, then the disintegration $\mathbb S=\{({\mathfrak A}_y,S_y):

Theorems & Definitions (22)

  • Definition 1.1
  • Proposition 1.2
  • Proposition 1.3
  • Theorem 1.4
  • Remark 1.5
  • Lemma 2.1
  • Lemma 2.2
  • Lemma 2.3
  • Definition 2.4
  • Remark 2.5
  • ...and 12 more