Mean-field models for non-Markovian epidemics on networks: from edge-based compartmental to pairwise models
arXiv:1611.04030
Abstract
This paper presents a novel extension of the edge-based compartmental model for epidemics with arbitrary distributions of transmission and recovery times. Using the message passing approach we also derive a new pairwise-like model for epidemics with Markovian transmission and an arbitrary recovery period. The new pairwise-like model allows one to formally prove that the message passing and edge-based compartmental models are equivalent in the case of Markovian transmission and arbitrary recovery processes. The edge-based and message passing models are conjectured to also be equivalent for arbitrary transmission processes; we show the first step of a full proof of this. The new pairwise-like model encompasses many existing well-known models that can be obtained by appropriate reductions. It is also amenable to a relatively straightforward numerical implementation. We test the theoretical results by comparing the numerical solutions of the various pairwise-like models to results based on explicit stochastic network simulations.
arXiv comment24 pages, 4 figures, submitted
References in corpus (9)
- The spread of epidemic disease on networks
- A modified Next Reaction Method for simulating chemical systems with time dependent propensities and delays
- A message passing approach for general epidemic models
- Edge-Based Compartmental Modeling for Infectious Disease Spread Part I: An Overview
- Second look at the spread of epidemics on networks
- Predicting the size and probability of epidemics in a population with heterogeneous infectiousness and susceptibility
- Simulating non-Markovian stochastic processes
- The relationships between message passing, pairwise, Kermack-McKendrick and stochastic SIR epidemic models
- Compact pairwise models for epidemics with multiple infectious stages on degree heterogeneous and clustered networks