Professor Paul Fearnhead
Distinguished Professor of StatisticsResearch Interests
- Bayesian & Computational statistics
- Anomaly and Change Detection
- Continuous-time Monte Carlo methods
Research Overview
I work at the interface of Statistics and Computation: aiming to develop new statistical methods and algorithms that are fast computationally and efficient statistically.
Recent work has included introducing pruning ideas that substantially speed-up state-of-the-art algorithms for offline and online change detection; and working on a new paradigm for Monte Carlo methods in Bayesian statistics based on simulating continous-time Markov processes.
Additional Information
Away from work I play bridge (mainly online in Warwickshire and Oxfordshire) and chess for Poulton Chess Club https://poultonchess.wordpress.com
Selected Publications
Fast Online Changepoint Detection via Functional Pruning CUSUM statistics
Romano, G., Eckley, I., Fearnhead, P., Rigaill, G. 15/10/2021 In: arxiv.org.
Journal article
The Zig-Zag Process and Super-Efficient Sampling for Bayesian Analysis of Big Data
Bierkens, J., Fearnhead, P., Roberts, G. 13/02/2019 In: Annals of Statistics. 47, 3, p. 1288-1320. 33 p.
Journal article
Optimal detection of changepoints with a linear computational cost
Killick, R., Fearnhead, P., Eckley, I. 17/10/2012 In: Journal of the American Statistical Association. 107, 500, p. 1590-1598. 9 p.
Journal article
Constructing summary statistics for approximate Bayesian computation: semi-automatic approximate Bayesian computation (with Discussion)
Fearnhead, P., Prangle, D. 06/2012 In: Journal of the Royal Statistical Society: Series B (Statistical Methodology). 74, 3, p. 419-474. 56 p.
Journal article
All Publications
Was that Change Real? Quantifying Uncertainty for Change Points
01/10/2021 → 30/09/2024
Research
DSI:STORi: Anomaly Detection for Real-time Condition Monitoring
01/08/2020 → 31/03/2024
Research
DSI: CoSInES: COmputational Statistical INference for Engineering and Security
01/10/2018 → 30/09/2023
Research
New Approaches to Bayesian Data Science: Tackling Challenges from the Health Sciences
01/04/2018 → 31/03/2023
Research
Statscale: Statistical Scalability for streaming data
01/06/2016 → 31/05/2023
Research
Intractable Likelihood: New Challenges From Modern Applications (iLike)
01/01/2013 → 30/06/2018
Research
Statistical Learning, STOR-i Centre for Doctoral Training
STOR-i Centre for Doctoral Training
Changepoints and Time Series, STOR-i Centre for Doctoral Training
Changepoints and Time Series, STOR-i Centre for Doctoral Training
STOR-i Centre for Doctoral Training
- Bayesian and Computational Statistics
- DSI - Foundations
- STOR-i Centre for Doctoral Training