Epidemics are often modelled using state-space models based on dynamical systems, observed through partial and noisy data. In this paper we develop stochastic extensions to the popular SEIR model with parameters evolving in time, in order to capture unknown influences of changing behaviors, public interventions, seasonal effects etc. Our models assign diffusion processes for the time-varying parameters, and our inferential procedure is based on the particle Markov Chain Monte Carlo algorithm, suitably adjusted to accommodate the features of this challenging nonlinear stochastic model. The performance of the proposed computational methods is validated on simulated data and the adopted model is applied to the 2009 A/H1N1 pandemic in England. In addition to estimating the trajectories of the effective contact rate, the methodology is applied in real time to provide evidence in related public health decisions.
View on arXiv