Bayesian Nonparametrics for Stochastic Epidemic Models

  title={Bayesian Nonparametrics for Stochastic Epidemic Models},
  author={Theodore Kypraios and Philip D. O’Neill},
  journal={Statistical Science},
The vast majority of models for the spread of communicable diseases are parametric in nature and involve underlying assumptions about how the disease spreads through a population. In this article we consider the use of Bayesian nonparametric approaches to analysing data from disease outbreaks. Specifically we focus on methods for estimating the infection process in simple models under the assumption that this process has an explicit time-dependence. 

Figures from this paper

Bayesian nonparametric inference for heterogeneously mixing infectious disease models
A flexible Bayesian nonparametric framework is proposed that avoids the need to make strict model assumptions about the infection process and enables a far more data-driven modeling approach for inferring the mechanisms governing transmission.
Emulation‐based inference for spatial infectious disease transmission models incorporating event time uncertainty
It is shown how both methods offer substantial computational efficiency gains over standard Bayesian MCMC‐based method, and can be used to infer the transmission of complex infectious disease systems, and also shows that importance sampling‐based methods tend to perform more satisfactorily.
Semiparametric Bayesian inference for the transmission dynamics of COVID-19 with a state-space model
Uniformly Ergodic Data-Augmented MCMC for Fitting the General Stochastic Epidemic Model to Incidence Data
A novel data-augmented MCMC algorithm for fast and exact Bayesian inference for the stochastic SIR model given discretely observed infection incidence counts and it is found that the method explores the high-dimensional latent space efficiently and scales to outbreaks with hundreds of thousands of individuals.
Modelling the reproductive power function
ABSTRACT This paper discusses methods of estimating the reproductive power and the accompanying survival function of communicable events, e.g. infectious disease transmission. The early stage of an
Optimal treatment allocations in space and time for on‐line control of an emerging infectious disease
A Bayesian on-line estimator of the optimal allocation strategy that combines simulation-optimization with Thompson sampling is derived and performs favourably in simulation experiments.
Does the data tell the true story? A modelling assessment of early COVID-19 pandemic suppression and mitigation strategies in Ghana
The results indicate that COVID-19 transmission reduced consistently in Ghana after the imposition of public health interventions—such as border restrictions, intra-city movement, quarantine and isolation—during the first phase of the pandemic from March to May 2020.
Bayesian inference for multistrain epidemics with application to ESCHERICHIA COLI O157:H7 in feedlot cattle
This work aims to demonstrate the importance of knowing the carrier and removal status of canine coronavirus in the context of food safety and infectious disease research.
Connecting and Contrasting the Bayes Factor and a Modified ROPE Procedure for Testing Interval Null Hypotheses
Abstract There has been strong recent interest in testing interval null hypotheses for improved scientific inference. For example, Lakens et al. and Lakens and Harms use this approach to study if


Bayesian inference for partially observed stochastic epidemics
The analysis of infectious disease data is usually complicated by the fact that real life epidemics are only partially observed. In particular, data concerning the process of infection are seldom
Bayesian non-parametric inference for stochastic epidemic models using Gaussian Processes
A Bayesian non-parametric approach using Gaussian Processes is developed, specifically to estimate the infection process, and it is illustrated that the methods can recover the true infection process quite well in practice.
Bayesian inference for stochastic epidemic models with time-inhomogeneous removal rates
A reversible jump MCMC algorithm is described that permits Bayesian inferences to be made on model parameters, particularly those associated with the step function, and applied to two datasets on outbreaks of smallpox and a respiratory disease.
Bayesian nonparametric inference for stochastic epidemic models
New methodology is developed which enables nonparametric estimation of the parameters which govern transmission within a Bayesian framework and works fairly well for the stochastic SIR model.
Inference in Epidemic Models without Likelihoods
This work investigates the performance of Markov Chain Monte Carlo and Sequential Monte Carlo algorithms for parameter inference, where the routines are based on approximate likelihoods generated from model simulations.
Estimating the Basic Reproductive Number in the General Epidemic Model with an Unknown Initial Number of Susceptible Individuals
Abstract.  In any epidemic, there may exist an unidentified subpopulation which might be naturally immune or isolated and who will not be involved in the transmission of the disease. Estimation of
Epidemic models and MCMC inference
Evidence confirming and demonstrating the importance of understanding the tail behaviour of proposals in importance sampling is presented, and a new algorithm, the Kernel Metropolis Hastings (KMH), is proposed as an approximate algorithm for low dimensional marginal inference in situations where the GIMH algorithm fails because of sticking.
Extracting the time-dependent transmission rate from infection data via solution of an inverse ODE problem
It is proved that almost any infection profile can be perfectly fitted by an SIR model with variable transmission rate, which clearly shows a serious danger of overfitting such transmission models.
Modelling and Bayesian analysis of the Abakaliki smallpox data.
Tractable nonparametric Bayesian inference in Poisson processes with Gaussian process intensities
This paper presents the first approach to Gaussian Cox processes in which it is possible to perform inference without introducing approximations or finitedimensional proxy distributions, and uses a generative model for Poisson data to enable tractable inference via Markov chain Monte Carlo.