An effective simulated annealing refined replica exchange Markov chain Monte Carlo algorithm for the infectious disease model of H1N1 influenza pandemic
Year: 2011
Type: article
Abstract: This paper is concerned with a computational algorithm for fitting a deterministic MSEIRS (immune-susceptible-exposed-infectious-recovered-susceptible) epidemic model for the transmission of influenza (H1N1) to mortality data. The model-fitting is carried out using a simulated annealing refined replica exchange Markov chain Monte Carlo algorithm. The effectiveness of the algorithm is illustrated using the triple wave data from five English towns collected during the 1918 ~ 1919 influenza pandemic. Numerical results show that the replica exchange (refined by simulated annealing) sampling technique is superior to other existing sampling techniques such as the Gibbs sampling technique, the Metropolis-Hastings sampling technique, the Multiple-try Metropolis technique for the Markov chain Monte Carlo computation. The algorithm presented in this paper has great promise to be used for carrying out some numerical computations of the current complex 2009 ∼ 2010 influenza pandemic.
Author: Zhang, Jiapu
Institution: Federation University
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2740541470