Computational methods for fitting stochastic epidemic models to data
Project description
Despite recent advances in the development of computational methods for fitting epidemic models to data, many of these methods work best in small-scale settings where the study population is not especially big or the models have relatively few parameters. There is a need to develop methods which are appropriate to large-scale settings. Furthermore, nearly all existing methods rely on parametric approaches (e.g. models based on specific underlying assumptions), but recent work has shown that Bayesian nonparametric approaches can be successfully adapted to this area. This project involves developing novel computationally efficient methods to fit both parametric and non-parametric models to data in situations where the existing methods are infeasible.
Project published references
- Xu, X., Kypraios, T. and O'Neill, P.D. ( 2016). Bayesian non-parametric inference for stochastic epidemic models using Gaussian Processes. Biostatistics. 17(4), 619-633
- Kypraios, T. and O'Neill, P.D. (2018) Bayesian nonparametrics for stochastic epidemic models. Statistical Science, 33(1): 44-56.
- Seymour, R.G., Kypraios, T., O'Neill, P.D. and Hagenaars, T.J. (2021) A Bayesian nonparametric analysis of the 2003 outbreak of highly pathogenic avian influenza in the Netherlands, To appear in Royal Statistical Society, Series C.
- Seymour, R. G., Kypraios, T., & O’Neill, P. D. (2022). Bayesian nonparametric inference for heterogeneously mixing infectious disease models. Proceedings of the National Academy of Sciences, 119(10), e2118425119.
More information
Full details of our Maths PhD
How to apply to the University of Nottingham