Maximum-entropy moment-closure for stochastic systems on networks

Tim Rogers

Research output: Contribution to journal › Article

22 Citations (Scopus)

Abstract

Moment-closure methods are popular tools to simplify the mathematical analysis of stochastic models defined on networks, in which high dimensional joint distributions are approximated (often by some heuristic argument) as functions of lower dimensional distributions. Whilst undoubtedly useful, several such methods suffer from issues of non-uniqueness and inconsistency. These problems are solved by an approach based on the maximisation of entropy, which is motivated, derived and implemented in this article. A series of numerical experiments are also presented, detailing the application of the method to the Susceptible-Infective-Recovered model of epidemics, as well as cautionary examples showing the sensitivity of moment-closure techniques in general.
Original languageEnglish
Article numberP05007
Number of pages20
JournalJournal of Statistical Mechanics-Theory and Experiment
Volume2011
Issue numberMay
DOIs
Publication statusPublished - May 2011

Fingerprint

Moment Closure
Maximum Entropy
Stochastic Systems
closures
entropy
moments
applications of mathematics
Nonuniqueness
Mathematical Analysis
Joint Distribution
Inconsistency
Stochastic Model
sensitivity
Simplify
High-dimensional
Entropy
Numerical Experiment
Heuristics
Series
Stochastic systems

Cite this

Maximum-entropy moment-closure for stochastic systems on networks. / Rogers, Tim.

In: Journal of Statistical Mechanics-Theory and Experiment, Vol. 2011, No. May, P05007, 05.2011.

Research output: Contribution to journal › Article

Rogers, T 2011, 'Maximum-entropy moment-closure for stochastic systems on networks', Journal of Statistical Mechanics-Theory and Experiment, vol. 2011, no. May, P05007. https://doi.org/10.1088/1742-5468/2011/05/P05007
Rogers T. Maximum-entropy moment-closure for stochastic systems on networks. Journal of Statistical Mechanics-Theory and Experiment. 2011 May;2011(May). P05007. https://doi.org/10.1088/1742-5468/2011/05/P05007
Rogers, Tim. / Maximum-entropy moment-closure for stochastic systems on networks. In: Journal of Statistical Mechanics-Theory and Experiment. 2011 ; Vol. 2011, No. May.
@article{cad29e04f06a4332bc68dacee76f6ac4,
title = "Maximum-entropy moment-closure for stochastic systems on networks",
abstract = "Moment-closure methods are popular tools to simplify the mathematical analysis of stochastic models defined on networks, in which high dimensional joint distributions are approximated (often by some heuristic argument) as functions of lower dimensional distributions. Whilst undoubtedly useful, several such methods suffer from issues of non-uniqueness and inconsistency. These problems are solved by an approach based on the maximisation of entropy, which is motivated, derived and implemented in this article. A series of numerical experiments are also presented, detailing the application of the method to the Susceptible-Infective-Recovered model of epidemics, as well as cautionary examples showing the sensitivity of moment-closure techniques in general.",
author = "Tim Rogers",
year = "2011",
month = "5",
doi = "10.1088/1742-5468/2011/05/P05007",
language = "English",
volume = "2011",
journal = "Journal of Statistical Mechanics-Theory and Experiment",
issn = "1742-5468",
publisher = "IOP Publishing",
number = "May",

}

TY - JOUR

T1 - Maximum-entropy moment-closure for stochastic systems on networks

AU - Rogers, Tim

PY - 2011/5

Y1 - 2011/5

N2 - Moment-closure methods are popular tools to simplify the mathematical analysis of stochastic models defined on networks, in which high dimensional joint distributions are approximated (often by some heuristic argument) as functions of lower dimensional distributions. Whilst undoubtedly useful, several such methods suffer from issues of non-uniqueness and inconsistency. These problems are solved by an approach based on the maximisation of entropy, which is motivated, derived and implemented in this article. A series of numerical experiments are also presented, detailing the application of the method to the Susceptible-Infective-Recovered model of epidemics, as well as cautionary examples showing the sensitivity of moment-closure techniques in general.

AB - Moment-closure methods are popular tools to simplify the mathematical analysis of stochastic models defined on networks, in which high dimensional joint distributions are approximated (often by some heuristic argument) as functions of lower dimensional distributions. Whilst undoubtedly useful, several such methods suffer from issues of non-uniqueness and inconsistency. These problems are solved by an approach based on the maximisation of entropy, which is motivated, derived and implemented in this article. A series of numerical experiments are also presented, detailing the application of the method to the Susceptible-Infective-Recovered model of epidemics, as well as cautionary examples showing the sensitivity of moment-closure techniques in general.

UR - http://www.scopus.com/inward/record.url?scp=79958045826&partnerID=8YFLogxK

UR - http://dx.doi.org/10.1088/1742-5468/2011/05/P05007

UR - http://arxiv.org/abs/1103.4980

U2 - 10.1088/1742-5468/2011/05/P05007

DO - 10.1088/1742-5468/2011/05/P05007

M3 - Article

VL - 2011

JO - Journal of Statistical Mechanics-Theory and Experiment

JF - Journal of Statistical Mechanics-Theory and Experiment

SN - 1742-5468

IS - May

M1 - P05007

ER -

Access to Document