We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Mathematics > Probability

Title: Computable Stochastic Processes

Abstract: The aim of this paper is to present an elementary computable theory of probability, random variables and stochastic processes. The probability theory is baed on existing approaches using valuations and lower integrals. Various approaches to random variables are discussed, including the approach based on completions in a Polish space. We apply the theory to the study of stochastic dynamical systems in discrete-time, and give a brief exposition of the Wiener process as a foundation for stochastic differential equations. The theory is based within the framework of type-two effectivity, so has an explicit direct link with Turing computation, and is expressed in a system of computable types and operations, so has a clean mathematical description.
Subjects: Probability (math.PR)
MSC classes: 60H10, 60A10, 03F60
Cite as: arXiv:1409.4667 [math.PR]
  (or arXiv:1409.4667v2 [math.PR] for this version)

Submission history

From: Pieter Collins [view email]
[v1] Mon, 15 Sep 2014 17:29:55 GMT (41kb)
[v2] Tue, 13 Oct 2015 10:02:41 GMT (47kb)

Link back to: arXiv, form interface, contact.