Stationary measures for randomly chosen maps
Burton RM, Keller G (1993)
Publication Type: Journal article, Original article
Publication year: 1993
Journal
Publisher: Springer Verlag (Germany)
Book Volume: 6
Pages Range: 1--16
Journal Issue: 1
DOI: 10.1007/BF01046765
Abstract
A Markov process on a compact metric space,X is given by random transformations.S is a finite set of continuous transformations ofX to itself. A random evolution onX is defined by lettingx inX evolve toT(x) forT inS with probability that depends onx andT but is independent of any other past measurable events. This type of model is often called a place dependent iterated function system. The transformations are assumed to have either monotone or contractive properties. Theorems are given to describe the number and types of ergodic invariant measures. Special emphasis is given to learning models and self-reinforcing random walks.
Authors with CRIS profile
Involved external institutions
United States (USA) (US)How to cite
APA:
Burton, R.M., & Keller, G. (1993). Stationary measures for randomly chosen maps. Journal of Theoretical Probability, 6(1), 1--16. https://dx.doi.org/10.1007/BF01046765
MLA:
Burton, Robert M., and Gerhard Keller. "Stationary measures for randomly chosen maps." Journal of Theoretical Probability 6.1 (1993): 1--16.
BibTeX: Download