Markov processes on Riesz spaces

DOI10.1007/S11117-011-0121-0MaRDI QIDQ1928535zbMATH OpenOpenAlexFDO

Authors Jessica J. Vardy, Bruce A. Watson

Publication date 3 January 2013

Published in Positivity (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1707.01054

conditional expectation independence Markov processes Riesz spaces

Linear operators on ordered spaces (47B60) Stopping times; optimal stopping problems; gambling theory (60G40) Martingales with discrete parameter (60G42) Generalizations of martingales (60G48)

Abstract: Measure-free discrete time stochastic processes in Riesz spaces were formulated and studied by Kuo, Labuschagne and Watson. Aspects relating martingales, stopping times, convergence of these processes as well as various decomposition were considered. Here we formulate and study Markov processes in a measure-free Riesz space setting.

Review text

Let \(E\) be a Riesz space with a weak order unit. A~positive order continuous projection \(T:E\to E\) with range \(R(T)\) a Dedekind complete Riesz subspace of \(E\) is called a conditional expectation if \(T(e)\) is a weak order unit of \(E\) for each weak order unit \(e\) of \(E\). Theorem. Let \(E\) be a Dedekind complete Riesz space with conditional expectation \(T\) and let \(e\) be a weak order unit which is invariant under \(T\). The band projection \(P\) and \(Q\) are \(T\)-conditionally independent if, and only if, \(TPTQ w = TPQ w = TQTP w\) for all \(w\in R(T)\). Next, the authors characterize independence of closed Riesz subspaces of a \(T\)-universally complete Riesz space in terms of conditional expectation operators. The following is one of the main results of this interesting paper. Theorem. Let \(T\) be a strictly positive conditional expectation on the \(T\)-universally complete Riesz space \(E\) with weak order unit \(e = Te\). Let \(\Lambda \) be a totally ordered index set. For \((X_i)_{i\in \Lambda}\subset E\), the following are equivalent: {\parindent=2.5 em\begin{itemize} \item[(i)] The process \((X_i)\) is a Markov process. \item [(ii)] For conditional expectations \(\mathbb T_u\) and \(T_v\) with \(R(\mathbb T_u) = \left\langle R(T),X_n:n\leq u \right\rangle\), \(R(T_v) = \left\langle R(T), X_v \right\rangle\), \(u < v \in \Lambda\), we have that \(\mathbb T_uT_v = T_uT_v\). \item [(iii)] For any \(s_m >\dots>s_1> t > t_n > \dots> t_1\) from \(\Lambda\) and \(P,Q\) band projections with \(Qe \in\left\langle R(T),X_{s_1},\dots,X_{s_m}\right\rangle\), we have that \(T_tQT_tPe = T_tQPe = T_tPQe = T_tPT_tQe\). \end{itemize}} There is a natural connection between sums of independent random variables and Markov processes. In Riesz spaces, this is given by the following result. Theorem. Let \(T\) be a strictly positive conditional expectation on the \(T\)-universally complete Riesz space \(E\) with order unit \(e = Te\). Let \((f_n)\) be a sequence in \(E\) which is \(T\)-conditionally independent. Then \((\sum_{k=1}^n)\) is a Markov process. The authors prove the following on the order convergence of independent summands. Theorem. Let \(T,E\) and \((f_n)\) be as above. If \(Tf_i= 0\) for all \(i\) and there exists \(g\in E\) such that \(T|\sum_{k=1}^n f_k|\leq g\) for all \(n\), then the partial sums of \(\sum_{k=1}^\infty f_k\) is order convergent. Finally, let \(T\) and \(E\) be as above and let \((f_n)\) be a Brownian motion in \(E\) with respect to \(T\). Then \((f_n)\) is a Markov process.

Recommendations

Cites work

Cited in

(27)

This page was built for publication: Markov processes on Riesz spaces

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1928535)