Revisiting Context-Tree Weighting for Bayesian Inference
View / Open Files
Authors
Papageorgiou, I
Kontoyiannis, I
Mertzanis, L
Panotopoulou, A
Skoularidou, M
Publication Date
2021Journal Title
IEEE International Symposium on Information Theory - Proceedings
Conference Name
2021 IEEE International Symposium on Information Theory (ISIT)
ISSN
2157-8095
ISBN
9781538682098
Publisher
IEEE
Volume
2021-July
Pages
2906-2911
Type
Conference Object
This Version
AM
Metadata
Show full item recordCitation
Papageorgiou, I., Kontoyiannis, I., Mertzanis, L., Panotopoulou, A., & Skoularidou, M. (2021). Revisiting Context-Tree Weighting for Bayesian Inference. IEEE International Symposium on Information Theory - Proceedings, 2021-July 2906-2911. https://doi.org/10.1109/ISIT45174.2021.9518189
Abstract
We revisit the statistical foundation of the celebrated context tree weighting (CTW) algorithm, and we develop a Bayesian modelling framework for the class of higher-order, variable-memory Markov chains, along with an associated collection of methodological tools for exact inference for discrete time series. In addition to deterministic algorithms that learn the a posteriori most likely models and compute their posterior probabilities, we introduce a family of variable-dimension Markov chain Monte Carlo samplers, facilitating further exploration of the posterior. The performance of the proposed methods in model selection, Markov order estimation and prediction is illustrated through simulation experiments and real-world applications.
Identifiers
External DOI: https://doi.org/10.1109/ISIT45174.2021.9518189
This record's URL: https://www.repository.cam.ac.uk/handle/1810/332891
Statistics
Total file downloads (since January 2020). For more information on metrics see the
IRUS guide.