My research interests lie in spatio-temporal statistics and large scale Bayesian inverse problems. I am also interested in Numerical Linear Algebra for large scale problems as well as Data Assimilation and their applications in statistics. I work under the supervision of Prof. Gavin Shaddick and Dr. Melina Freitag. I am part of the Bath Statistics group and Bath Numerical Analysis group.

PhD Project Description

Air pollution poses a significant threat to global health and has been associated with a range of adverse health effects, including cardiovascular and respiratory diseases in addition to some cancers. It is vital that the subsequent risks, trends and consequences of air pollution are monitored and modelled to develop effective environmental and public health policy to lessen the burden of air pollution. 

Estimating the health burden associated with air pollution requires accurate estimates of the exposures experienced by populations. One of the primary sources for information of the levels of air pollution is ground monitoring. Although there is an increasing number of monitors, the majority are located in North America, Western Europe, China and India, in many areas ground monitoring remains sparse. One of the challenges in assessing the global effects of air pollution is to obtain estimates of air pollution where there is no ground monitoring information. There is a need therefore to supplement ground monitoring data with information from other sources, such as estimates of air pollution from satellite remote sensing, chemical transport models, and information on land use and population. 

The primary focus of my research with Professor Gavin Shaddick is the development of Bayesian hierarchical models for integrating data from multiple sources in order to produce an accurate, high resolution set of exposures to air pollution over space and time, accompanied with measures of uncertainty. Due to both the complexity of the models and the high-dimensional data, traditional methods for statistical inference of these models, such as Markov Chain Monte Carlo (MCMC), can present some computational challenges. Therefore, we make use of recently developed techniques that perform approximate Bayesian inference such as integrated nested Laplace approximations (INLA), which have been developed as a computationally attractive alternative to MCMC and do not require sampling to be performed. Ultimately this research works towards more accurate estimates of the burden of disease attributable to air pollution.


Morrison, K. T., Shaddick, G., Thomas, M. L., Buckeridge, D. L. & Henderson, S. B. (2016). Improving forecast accuracy in environmental public health surveillance: a spatio-temporal approach. Manuscript submitted for publication.

Cohen, A., Brauer, M., Burnett, R., Anderson, H. R., Frostad, J., Estep, K., Balakrishnan, K., Brunekreef, B., Dandona, L., Dandona, R., Feigin, V., Freedman, G., Hubbell, B., Haidong, K., Jobling, A., Knibbs, L., Liu, Y., Martin, R., Morawska, L., Pope, C., A., Shin, H., Straif, K., Shaddick, G., Thomas, M. L., van Dingenen, R., van Donkelaar, A., Vos, T., Murray, C. J. L. and Forouzanfar, M. H. (2017). The Global Burden of Disease attributable to ambient air pollution: Estimates of current burden and 25-year trends from the GBD2015 study. Accepted: To appear in The Lancet

Shaddick, G., Thomas M. L., Jobling A., Brauer M., van Donkelaar, A., Burnett, R., Chang H., Cohen A., Van Dingenen, R., Dora, C., Gumy, S., Liu, Y., Martin, R., Waller, L. A., West, J., Zidek, J. V. and Prüss-Ustün, A. (2017). Data Integration Model for Air Quality: A Hierarchical Approach to the Global Estimation of Exposures to Ambient Air Pollution To appear in Journal of the Royal Statistical Society: Series C (Applied Statistics).

Forouzanfar, M. H., Afshin, A., Alexander, L. T., Anderson, H. R., ..., Thomas, M. L., et al. (2016) Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks in 188 countries, 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015. The Lancet

Tillett, W., Shaddick, G., Jobling, A., Askari, A., Cooper, A., Creamer, P., Clunie, G., Helliwell, P. S., James, J., Kay, L., Korendowych, E., Lane S., Packham, J., Shaban, R., Thomas, M. L., Williamson, L. and McHugh, N. (2016). Effect of anti-TNF and conventional synthetic disease-modifying anti-rheumatic drug treatment on work disability and clinical outcome in a multicentre observational cohort study of psoriatic arthritis. Rheumatology.

Tillett, W., Shaddick, G., Jobling, A., Thomas, M., Korendowych, E., & McHugh, N. (2015). OP0001 Work Disability After Initiation of Anti-TNF and DMARD Treatment in Psoriatic Arthritis; Investigator LED, UK, Multicentre Observational Cohort Study (LOPAS II). Annals of the Rheumatic Diseases, 74(Suppl 2), 64-64.