Stopped processes and Doob's optional sampling theorem
Jacobus J. Grobler, Christopher M. Schwanke

TL;DR
This paper extends the theory of stopped processes by defining a new spectral measure-based stopping element, generalizes Doob's optional sampling theorem without the need for the Doob-Meyer property, and introduces unbounded order convergence for a more comprehensive understanding of convergence in stochastic processes.
Contribution
It introduces a spectral measure approach to defining stopped processes, removes the Doob-Meyer assumption in the optional sampling theorem, and employs unbounded order convergence for a broader convergence framework.
Findings
Defined the stopping element $X_S$ via spectral measure integration.
Provided a new proof of Doob's optional sampling theorem without Doob-Meyer assumption.
Demonstrated convergence properties of uniformly integrable sequences under unbounded order convergence.
Abstract
Using the spectral measure of the stopping time we define the stopping element as a Daniell integral for an adapted stochastic process that is a Daniell summable vector-valued function. This is an extension of the definition previously given for right-order-continuous sub-martingales with the Doob-Meyer decomposition property. The more general definition of necessitates a new proof of Doob's optional sampling theorem, because the definition given earlier for sub-martingales implicitly used Doob's theorem applied to martingales. We provide such a proof, thus removing the heretofore necessary assumption of the Doob-Meyer decomposition property in the result. Another advancement presented in this paper is our use of unbounded order convergence, which properly characterizes the notion…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
