Skip to main content

Research Repository

Advanced Search

How to correctly fit an SIR model to data from an SEIR model?

KhudaBukhsh, Wasiur R.; Rempała, Grzegorz A.

How to correctly fit an SIR model to data from an SEIR model? Thumbnail


Authors

Grzegorz A. Rempała



Abstract

In epidemiology, realistic disease dynamics often require Susceptible-Exposed-Infected-Recovered (SEIR)-like models because they account for incubation periods before individuals become infectious. However, for the sake of analytical tractability, simpler Susceptible-Infected-Recovered (SIR) models are commonly used, despite their lack of biological realism. Bridging these models is crucial for accurately estimating parameters and fitting models to observed data, particularly in population-level studies of infectious diseases. This paper investigates stochastic versions of the SEIR and SIR frameworks and demonstrates that the SEIR model can be effectively approximated by a SIR model with time-dependent infection and recovery rates. The validity of this approximation is supported by the derivation of a large-population Functional Law of Large Numbers (FLLN) limit and a finite-population concentration inequality. To apply this approximation in practice, the paper introduces a parameter inference methodology based on the Dynamic Survival Analysis (DSA) survival analysis framework. This method enables the fitting of the SIR model to data simulated from the more complex SEIR dynamics, as illustrated through simulated experiments.

Citation

KhudaBukhsh, W. R., & Rempała, G. A. (2024). How to correctly fit an SIR model to data from an SEIR model?. Mathematical Biosciences, 375, Article 109265. https://doi.org/10.1016/j.mbs.2024.109265

Journal Article Type Article
Acceptance Date Jul 26, 2024
Online Publication Date Jul 30, 2024
Publication Date 2024-09
Deposit Date Jul 27, 2024
Publicly Available Date Jul 31, 2025
Journal Mathematical Biosciences
Print ISSN 0025-5564
Electronic ISSN 1879-3134
Publisher Elsevier
Peer Reviewed Peer Reviewed
Volume 375
Article Number 109265
DOI https://doi.org/10.1016/j.mbs.2024.109265
Public URL https://nottingham-repository.worktribe.com/output/37606011
Additional Information This article is maintained by: Elsevier; Article Title: How to correctly fit an SIR model to data from an SEIR model?; Journal Title: Mathematical Biosciences; CrossRef DOI link to publisher maintained version: https://doi.org/10.1016/j.mbs.2024.109265; Content Type: article; Copyright: © 2024 The Authors. Published by Elsevier Inc.

Files





You might also like



Downloadable Citations