University of Oxford, Department of Statistics Pride logo
Professor Geoff Nicholls
Associate Professor of Statistics

New collaborations welcome!

Biographical Sketch

Professor Geoff Nicholls, B.Sc. (Physics, Canterbury, New Zealand), MA, PhD (DAMTP, Cambridge, UK), researches Bayesian statistical methodology and lectures in statistics in the Department. He teaches probability, statistics and applied mathematics in St Peter's college. Geoff joined the Statistics Department in 2005 from the Mathematics Department of the University of Auckland in New Zealand. He was Head of Department 2012-2015.  

Research Interests

  • Applied Bayesian inference
  • Bayesian statistical methods
  • Computational statistics
  • Monte Carlo and Variational methods

Geoff is working on Bayesian statistical inference for problems with computationally demanding prior and likelihood evaluations. Research is driven by problems presented by scientists and scholars working in a range of application areas, including Geoscience, Linguistics, History and Sociology, Genetics and Archaeology. Solving these problems motivates new statistical models, well adapted computational tools and novel statistical methodology.

 

Publications

Carmona, C. and Nicholls, G. (2020) “Semi-Modular Inference: enhanced learning in multi-modular models by tempering the influence of components”, in Proceedings of Machine Learning Research, pp. 4226–4235.
Cui, T. et al. (2019) “Using parallel Markov chain Monte Carlo to quantify uncertainties in geothermal reservoir calibration”, International Journal for Uncertainty Quantification, 9(3), pp. 295–310.
Xing, H., Nicholls, G. and Lee, J. (2019) “Calibrated approximate Bayesian inference”, in Proceedings of Machine Learning Research. Journal of Machine Learning Research, pp. 6912–6920.
Rodrigues, E. et al. (2019) “Using a non-homogeneous Poisson model with spatial anisotropy and change-points to study air pollution data”, Environmental and Ecological Statistics, 26(2), pp. 153–184.
Moores, M. et al. (2018) “Scalable Bayesian inference for the inverse temperature of a hidden Potts model”, Bayesian Analysis [Preprint].
Roeling, M. and Nicholls, G. (2018) “Stochastic Block Models as an Unsupervised Approach to Detect Botnet-Infected Clusters in Networked Data”, in Data Science for Cyber-Security. World Scientific Publishing, pp. 161–178.
Lee, J., Nicholls, G. and Ryder, R. (2018) “Calibration procedures for approximate Bayesian credible sets”, arXiv.
Kelly, L. and Nicholls, G. (2017) “Lateral transfer in Stochastic Dollo models”, Annals of Applied Statistics, 11(2), pp. 1146–1168.
Styring, A. et al. (2017) “Isotope evidence for agricultural extensification reveals how the world’s first cities were fed”, Nature Plants, 3, p. 17076.
Matechou, E. et al. (2016) “Bayesian analysis of Jolly-Seber type models: Incorporating heterogeneity in arrival and departure”, Environmental and Ecological Statistics, 23(4), pp. 531–547.

Contact Details

College affiliation: Fellow and Tutor at St Peter's College

Office: 1.12 

Research Students

Current DPhil students: 

Sitong Liu Generalised Bayesian inference
Laura Battaglia Bayes methods for misspecified models
Holly Li Scalable Statistical inference for partial orders
Shiyi Sun New variational methods for Bayesian inference

Recent Graduates - DPhil:

Jessie Jiang (2024) Statistical inference for partial orders
Schyan Zafar (2024) Bayesian inference for multivariate time series 
Chris Carmona (2023) Bayesian Semi-Modular Inference
Lorenzo Pacchiardi (2022) Statistical inference in generative models (with R Dutta)
Hanwen Xing (2022) Diagnostic Methods for Bayesian Inference