Epidemic thresholds identification of susceptible-infected-recovered model based on the Eigen Microstate

Ning-Ning Wang,Shui-Han Qiu,Xiao Wen Zhong,Zeng-Ru Di
DOI: https://doi.org/10.1016/j.amc.2023.127924
2023-01-01
Abstract:Epidemic threshold estimation plays a critical role in prevention strategies. Besides the-oretical methods, some numerical methods estimate the threshold through the scale di-vergence of outbreak size near critical state. But the bimodal distribution of outbreak size may make the estimated threshold higher. In this paper, we extend the Eigen Microstate (EM) method to the susceptible-infected-recovered (SIR) epidemic model and devise the eigen measure to estimate the threshold using the scale divergence of singular values. Compared with the susceptibility and variability measure, the estimated threshold of eigen measure is closer to the theoretical results of the heterogeneous mean-field (HMF) and the quenched mean-field (QMF). In addition, better than the assortativity coefficient, network mean degree or dimension has a more significant effect on estimation accuracy. Under complex epidemic mechanisms, such as periodic time-varying networks and migration be-tween networked populations, the threshold estimation of QMF models indicates that the eigen measure is more stable and accurate than the other two measures. Combined with renormalization group theory, our method may have practical significance in megalopolis epidemic warning.(c) 2023 Elsevier Inc. All rights reserved.
What problem does this paper attempt to address?