Dr Murray Pollock

Director of Statistics

Background

I am currently the Director of Statistics within the School of Mathematics, Statistics and Physics at Newcastle University, holding additional responsibility as the Director for the CDT in Cloud Computing and Big Data (run jointly with the School of Computing). I concurrently hold an appointment as a Turing Fellow based within the Alan Turing Institute at the British Library (the United Kingdom’s national institute for data science and artificial intelligence), and as an Honorary Associate Professor based within the Department of Statistics at the University of Warwick. I am also a trustee and council member of the Royal Statistical Society, an Associate Editor of "Bayesian Analysis", and act as External for the University of Essex. My research is primarily in computational statistics and data science, and is broadly concerned with addressing practical constraints that arise in modern inferential application settings (for instance algorithmic scalability with data size, working under data privacy constraints, and methodological design in parallel and distributed computing environments).

Research

My current research interests include cryptography, machine learning, Monte Carlo methodology, perfect simulation, risk modelling and stochastic differential equations.

Teaching

For the 2020/21 academic year I am teaching MAS8951 Modern Bayesian Inference

Publications

Pollock M, Fearnhead P, Johansen AM, Roberts GO. Quasi-stationary Monte Carlo and the ScaLE algorithm. Journal of the Royal Statistical Society. Series B: Statistical Methodology 2020, 82(5), 1167-1221.
Wang AQ, Pollock M, Roberts GO, Steinsaltz D. Regeneration-enriched Markov processes with application to Monte Carlo. arXiv 2020, arXiv:1910.05037v3.
Fearnhead P, Bierkens J, Pollock M, Roberts GO. Piecewise deterministic Markov processes for continuous-time Monte Carlo. Statistical Science 2018, 33(3), 386-412.
Pollock M, Johansen AM, Roberts GO. On the exact and e-strong simulation of (jump) diffusions. Bernoulli 2016, 22(2), 794-856.
Mider M, Jenkins PA, Pollock M, Roberts GO. The Computational Cost of Blocking for Sampling Discretely Observed Diffusions. Methodology and Computing in Applied Probability 2022, Epub ahead of print.
Hodgson J, Johansen AM, Pollock M. Unbiased Simulation of Rare Events in Continuous Time. Methodology and Computing in Applied Probability 2022, 24, 2123-2148.
Dai H, Pollock M, Roberts G. Monte Carlo fusion. Journal of Applied Probability 2019, 56(1), 174-191.
Pollock M. On the exact simulation of (jump) diffusion bridges. In: 2015 Winter Simulation Conference (WSC). 2015, Huntington Beach, CA, USA: Institute of Electrical and Electronics Engineers Inc.