Statistical Science


Dr Alexandros Beskos

PositionReader in Statistics
Phone (external)02076798352
Phone (internal)48352
Personal webpagehttps://www.ucl.ac.uk/statistics/people/alexandrosbeskos
ThemesComputational Statistics, Stochastic Modelling of Complex Systems

* @ucl.ac.uk

Alexandros Beskos…

Biographical Details

Since 2015: Reader (Associate Professor) in Statistics, Department of Statistical Science, UCL.

Since 2017: Research Fellow of the Alan Turing Institute for Data Science.

2020 (1st half): Visiting Associate Professor, Yale-NUS College, Singapore.

2014-2015: Visiting Researcher, Engineering Department, Signal Processing & Communications Laboratory, University of Cambridge.

2013-2014: Associate Professor, Department of Statistics & Applied Probability, National University of Singapore, Singapore.

2012-2013: Senior Lecturer in Statistics, Department of Statistical Science, UCL. 

2008-2012: Lecturer in Statistics, Department of Statistical Science, UCL. 

2005-2008: Post-Doc, Mathematics Institute and Department of Statistics, University of Warwick. 

2002-2005: PhD, Lancaster University, supervisor Professor Gareth Roberts.

Research Interests

Bayesian Statistics, Computational Statistics, Machine Learning, Stochastic Volatility, Financial Statistics, Inverse Problems, Filtering, Econometrics, Biostatistics, Graphical Models, Models in Epigenetics, Population Models.

Selected publications

  • Efficient Sequential Monte Carlo Algorithms for Integrated Population Models, with Axel Finke, Ruth King, Petros Dellaportas. Journal of Agricultural, Biological, and Environmental Statistics, 24(2019), 204-224.
  • Optimization Based Methods for Partially Observed Chaotic Systems, with Daniel Paulin, Ajay Jasra, Dan Crisan. Journal of Foundations of Computational Mathematics, 19(2019), 485-559.
  • Particle Filtering for Stochastic Navier-Stokes Signal Observed with Linear Additive Noise, with Francesc Llopis, Nikolas Kantas, Ajay Jasra, SIAM Journal of Scientific Computing, 40(2018), A1544-A1565.                                         
  • Bridging Tree for Posterior Inference on Ancestral Recombination Graphs, with Kari Heine, Ajay Jasra, David Balding, Maria De Iorio. Proceedings of Royal Society A, 474(2018), 20180568.
  • Multilevel Sequential Monte Carlo with Dimension-Independent Likelihood-Informed Proposals, with Ajay Jasra, Kody Law, Youssef Marzouk, Yan Zhou. SIAM/ASA Journal of Uncertainty Quantification, 6(2018), 762-786.
  • Asymptotic Analysis of the Random-Walk Metropolis Algorithm on Ridged Densities, with Gareth Roberts, Alex Thiery, Natesh Pillai. The Annals of Applied Probability, 28(2018), 2966-3001.
  • On Concentration Properties of Partially Observed Chaotic Systems, with Daniel Paulin, Ajay Jasra, Dan Crisan. Advances in Applied Probability, 50(2018), 440-479.
  • A Stable Particle Filter for a Class of High-Dimensional State-Space Models, with Dan Crisan, Ajay Jasra, Kengo Kamatani, Yan Zhou. Advances in Applied Probability, 49(2017), 1-25.
  • Geometric MCMC for Infinite-Dimensional Inverse Problems, with Mark Girolami, Shiwei Lan, Farell, Andrew Stuart. Journal of Computational Physics, 335(2017), 327-351.
  • Multilevel Sequential Monte-Carlo Samplers, with Ajay Jasra, Kody Law, Raul Tempone, Yan Zhou. Stochastic Processes and their Applications, 127(2017), 1417-1440.
  • Natural Hedge of a Gas-Fired Power Plant, with Xiaojia Guo, Afzal Siddiqui. Computational Management Science, 13(2016), 63-86.
  • On the Convergence of Adaptive Sequential Monte Carlo Methods, with Ajay Jasra, Nikolas Kantas, Alex Thiery. The Annals of Applied Probability, 26(2016), 1111-1146.
  • Bayesian Inference for Partially Observed Stochastic Differential Equations Driven by Fractional Brownian Motion, with Joseph Dureau, Konstantinos Kalogeropoulos. Biometrika,  102(2015), 809-827.
  • Bayesian inference for Duplication-Mutation with Complementarity Network models, with Ajay Jasra, Adam Persing, Kari Heine, Maria De Iorio. Journal of Computational Biology, 22, 11(2015), 1025-1033.
  • Sequential Monte-Carlo methods for high-dimensional Inverse Problems: a case study for the Navier-Stokes equations, with Nikolas Kantas, Ajay Jasra. SIAM/ASA Journal of Uncertainty Quantification, 2, 1(2014), 464-489
  • On the stability of Sequential Monte-Carlo methods in high dimensions, with Dan Crisan, Ajay Jasra. The Annals of Applied Probability, 46, 4(2014), 1396-1445.
  • Advanced MCMC methods for sampling on diffusion pathspace, with Kostas Kalogeropoulos, Erik Pazos. Stochastic Processes and their Applications, 123, 4(2013), 1415-1453. 
  • Optimal tuning of the Hybrid Monte-Carlo, with Natesh Pillai, Gareth Roberts, Jesus Maria Sanz-Serna, Andrew Stuart. Bernoulli, 19, 5A(2013), 1501-1534.
  • \epsilon-strong simulation of the Brownian path, with Stefano Peluchetti, Gareth Roberts. Bernoulli, 18, 4(2012), 1223-1248.
  • Hybrid Monte-Carlo on Hilbert spaces, with Frank Pinski, Jesus Maria Sanz-Serna, Andrew Stuart. Stochastic Processes and their Applications, 121, 10(2011), 2201-2230. 
  • Optimal scalings for local Metropolis-Hastings chains on nonproduct targets in high dimensions, with Gareth Roberts, Andrew Stuart. The Annals of Applied Probability, 19, 3(2009), 863-898.
  • Monte-Carlo maximum likelihood estimation for discretely observed diffusion processes, with Omiros Papaspiliopoulos, Gareth Roberts. The Annals of Statistics, 37, 1(2009), 223-245.
  • Retrospective exact simulation of diffusion sample paths with applications, with Omiros Papaspiliopoulos, Gareth Roberts. Bernoulli, 12, 6(2006), 1077-1098.
  • Exact simulation of diffusions, with Gareth Roberts. The Annals of Applied Probability, 15, 4(2005), 2422-2444. 
  • Exact and computationally efficient likelihood-based estimation for discretely observed diffusion processes (with discussion), with Omiros Papaspiliopoulos, Gareth Roberts, Paul Fearnhead. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 68, 3(2006), 333-382.

* For an up-to-date complete list of publications, see my google scholar webpage: