Inference in Stochastic Epidemic Models via Multinomial Approximations
This provides a computationally efficient inference method for epidemiologists modeling outbreaks with incomplete data, though it is incremental as it builds on existing compartmental models.
The authors tackled inference in stochastic epidemic models with partial or missing data by introducing a method using recursive multinomial approximations to avoid likelihood intractability, demonstrating accuracy on Ebola and COVID-19 data without requiring forward simulation or tuning parameters.
We introduce a new method for inference in stochastic epidemic models which uses recursive multinomial approximations to integrate over unobserved variables and thus circumvent likelihood intractability. The method is applicable to a class of discrete-time, finite-population compartmental models with partial, randomly under-reported or missing count observations. In contrast to state-of-the-art alternatives such as Approximate Bayesian Computation techniques, no forward simulation of the model is required and there are no tuning parameters. Evaluating the approximate marginal likelihood of model parameters is achieved through a computationally simple filtering recursion. The accuracy of the approximation is demonstrated through analysis of real and simulated data using a model of the 1995 Ebola outbreak in the Democratic Republic of Congo. We show how the method can be embedded within a Sequential Monte Carlo approach to estimating the time-varying reproduction number of COVID-19 in Wuhan, China, recently published by Kucharski et al. 2020.