Dr Antony Overstall

Associate Professor in Statistics


Mathematical Sciences
University of Southampton
Southampton
SO17 1BJ

Tel: +44 (0) 23 8059 2724
E-mail: A.M.Overstall@soton.ac.uk
Photograph of Antony Overstall

Research Interests

My research interests lie in the areas of
  1. optimal experimental design;
  2. the analysis of categorical data.

Optimal experimental design involves allocating the (often, limited) resources of a physical experiment to, essentially, maximise the amount of "information" that the experiment provides. Increasingly, I am interested in phenomena which are explained by a model which is implemented by some computationally expensive computer code.

I am also interested in categorical data analysis, particularly incomplete contingency tables. The main application of these has been estimating the number of people who inject drugs in England and Scotland.

Working papers

Invited Talks & Seminars

  1. Robust Bayesian Design of Experiments for Calibration of Mathematical Models. Invited talk. SIAM Conference on Computational Science and Engineering (Virtual). 5th March 2021.
  2. Challenges of estimating human population sizes. Invited talk. International Biometric Society: Presidential Address and Annual General Meeting (Virtual). 11th November 2020.
  3. Bayesian prediction for physical models with application to the optimisation of the synthesis of pharmaceutical products using chemical kinetics. Statistics Seminar. School of Mathematical Sciences and Actuarial Science, University of Kent, Canterbury, UK. 23rd January 2020.
  4. Bayesian Optimal Design for Ordinary Differential Equation Models.Invited Talk. 12th International Conference of the ERCIM WG on Computational and Methodological Statistics, University of London, UK. 16th December 2019
  5. Bayesian design for intractable likelihood models. Invited Talk. 10th International Workshop on Simulation and Statistics, University of Salzburg, Salzburg, Austria. 5th September 2019.
  6. Bayesian design for physical models using computer experiments. Invited Talk. Spring Research Conference on Statistics in Industry and Technology. Virginia Tech, Blacksburg, Virginia, USA. 22nd May 2019.
  7. Bayesian design for physical models using computer experiments. Invited Talk. 5th International Conference on Design of Experiments. University of Memphis, Memphis, Tennessee, USA. 20th May 2019.
  8. Bayesian design for intractable likelihood models using computer experiments. Statistics Seminar. Department of Applied Statistics, Johannes Kepler University, Linz, Austria. 4th April 2019.
  9. Bayesian design for intractable likelihood models. Statistics Seminar. Department of Mathematics, King's College, London, UK. 28th March 2019.
  10. Bayesian prediction for physical models with application to the optimisation of the synthesis of pharmaceutical products using chemical kinetics. Statistics Seminar. School of Mathematics, University of Edinburgh, Edinburgh, UK. 16th November 2018.
  11. Bayesian design for intractable models. Invited Talk. Joint Statistical Meetings, Vancouver, British Columbia, Canada. 1st August 2018.
  12. Bayesian design for intractable models. Invited Talk. International Symposium on Business and Industrial Statistics (ISBIS), University of Piraeus, Piraeus, Greece. 5th July 2018.
  13. Bayesian design for intractable models. Invited Talk. 2nd International Conference on Econometrics and Statistics (ECOSTAT), City University of Hong Kong, Hong Kong. 20th June 2018.
  14. Experiments, statistics and uncertainty quantification. Invited Talk. London Mathematical Society Summer School, University of Manchester, UK. 21st July 2017.
  15. Optimal design for supersaturated experiments. Invited Talk. 4th Greco-Italian Meeting on Statistics, Università degli Studi di Firenze, Florence, Italy. 4th July 2017.
  16. Bayesian optimal design of experiments for models based on systems of ODEs. Invited Talk. Statistical Perspectives on Uncertainty Quantification, Georgia Institute of Technology, Atlanta, Georgia, USA. 30th May 2017.
  17. Bayesian optimal design of experiments using normal-based approximations to posterior quantities. Invited Talk. Conference on Experimental Design and Analysis, Academia Sinica, Taipei, Taiwan. 17th December 2016.
  18. Bayesian Optimal Design for Ordinary Differential Equation Models. Statistics Seminar. Escuela de Arquitectura, Universidad Castilla La Mancha, Toledo, Spain. 10th November 2016.
  19. Bayesian Optimal Design for Ordinary Differential Equation Models. Statistics Seminar. School of Mathematics, University of Manchester, Manchester, UK. 3rd November 2016.
  20. Bayesian optimal design of experiments for models based on systems of ODEs. Invited Talk. SIAM Conference on Uncertainty Quantification, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland. 8th April 2016.
  21. Bayesian optimal design of experiments for models based on systems of ODEs. Invited Talk. International Association for Statistical Computing Asian Region Section Conference, National University of Singapore, Singapore. 19th December 2015.
  22. Bayesian optimal design for ordinary differential equation models. Invited Talk. Designed Experiments: Recent Advances in Methods and Applications, University of Technology Sydney, Sydney, New South Wales, Australia. 14th December 2015.
  23. Approximate Coordinate Exchange (ACE) Algorithm for Bayesian Optimal Design. Invited Talk. Bayesian Optimal Design of Experiments (BODE) Workshop, Queensland University of Technology, Brisbane, Queensland, Australia. 10th December 2015.
  24. Bayesian optimal design of experiments for models based on systems of ODEs. Invited Talk. Bayes on the Beach, Surfers Paradise, Queensland, Australia. 9th December 2015.
  25. Bayesian Optimal Design for Ordinary Differential Equation Models. Statistics Seminar. School of Mathematics & Statistics, Newcastle University, Newcastle, UK. 30th October 2015.
  26. Estimating the prevalence of injecting drug use in Scotland using capture-recapture and partially observed contingency tables. Invited Talk. World Statistics Congress, Rio de Janeiro, Brazil. 28th July 2015.
  27. Bayesian optimal design of experiments for models based on systems of ODEs. Invited Talk. 3rd Greco-Italian Meeting on Statistics, Athens, Greece. 26th June 2015.
  28. Bayesian optimal design for ordinary differential equation models. Statistics Seminar. Biomathematics and Statistics Scotland (BioSS), James Hutton Institute, Aberdeen, UK. 2nd June 2015.
  29. Bayesian optimal design for computationally expensive models. Statistics Seminar. Mathematics Sciences, University of Southampton, Southampton, UK. 13th November 2014.
  30. Bayesian Inference and Optimal Design for Differential Equation Models with Application to Chemical Kinetics. Invited Talk. Joint Statistical Meetings, Boston, Massachusetts, USA. 5th August 2014.
  31. Bayesian optimal design for estimating the physical parameters of differential equation models. Statistics Seminar. Department of Statistics, University of Oxford, Oxford, UK. 4th June 2014.
  32. The approximate coordinate exchange algorithm for Bayesian optimal experimental design. Statistics Seminar. Department of Mathematics and Statistics, Lancaster University, Lancaster, UK. 28th March 2014.
  33. A Strategy for Bayesian Inference for Computationally Expensive Models with Application to the Estimation of Stem Cell Properties. Statistics Seminar. CSIRO/Mathematics Sciences, Australian National University, Canberra, Australia. 5th July 2013.
  34. Optimal Bayesian Design using Gaussian Process Emulators. Invited Talk. Spring Research Conference on Statistics in Industry and Technology. UCLA, Los Angeles, California, USA. 22nd June 2013.
  35. A Strategy for Bayesian Inference for Computationally Expensive Models with Application to the Estimation of Stem Cell Properties. Statistics Seminar. Mathematics Sciences, University of Southampton, Southampton, UK. 6th December 2012.
  36. Bayesian Lightweight Emulation of a Multivariate Simulator for a Humanitarian Relief Scenario. Invited Talk. Spring Research Conference on Statistics in Industry and Technology. Northwestern University, Evanston, Illinois, USA. 23rd June 2011.

Contributed Talks & Posters

  1. Decision-theoretic frequentist optimal design of experiments. Contributed Talk. Joint Statistical Meetings, Chicago, Illinois, USA. 31st July 2016.
  2. Bayesian optimal design for estimating physical parameters of computer models with application to measuring properties of human placentas. Poster. DAE2015: Design and Analysis of Experiments Conference, SAS World Headquarters, Cary, North Carolina, USA, 5th March 2015.
  3. Bayesian optimal design for estimating the physical parameters of computer models based on a system of ordinary differential equations. Contributed Talk. Uncertainty in Computer Models Conference. University of Sheffield, Sheffield, UK. 29th July 2014.
  4. Bayesian Inference for Computationally Expensive Models with Application to the Estimation of Stem Cell Properties. Contributed Talk. Western and North American Region of the International Biometrics Society Conference. UCLA, Los Angeles, California, USA. 17th June 2013.
  5. Design space production for chemical kinetics models. Poster. Turing Gateway to Mathematics: Industrial Statistics Day, Isaac Newton Institute for Mathematical Sciences, Cambridge, UK. 25th March 2013.
  6. Emulating the Likelihood/Posterior Function via Sequential Design. Contributed Talk. Joint Statistical Meetings. South Beach, Miami, Florida, USA. 3rd August 2011.

PhD Students


Former PhD Students


Software


Miscellaneous


Publications

  1. Overstall, A.M. (2021). Properties of using Fisher information gain for Bayesian design of experiments. Journal of Statistical Planning and Inference (To appear)
  2. Overstall, A.M. & McGree J.M. (2021). Bayesian decision-theoretic design of experiments under an alternative model. Bayesian Analysis (To appear)
  3. Cruyff, M., Overstall, A.M., Papathomas, M, & McCrea, R. (2021). Multiple System Estimation of Victims of Human Trafficking: Model Assessment and Selection. Crime & Delinquency (To appear)
  4. Sharifi Far, S., King, R., Bird, S., Overstall, A.M., Worthington, H., Jewell, N. (2021). Multiple Systems Estimation for Modern Slavery: Robustness of List Omission and Combination. Crime & Delinquency (To appear)
  5. Senarathne, J.S.G., Overstall, A.M. & McGree J.M. (2020). Bayesian adaptive N-of-1 trials for estimating population and individual treatment effects. Statistics in Medicine 39(29) 4499-4518
  6. Overstall, A.M., Woods, D.C. & Adamou, M. (2020). acebayes: An R Package for Bayesian Optimal Design of Experiments via Approximate Coordinate Exchange. Journal of Statistical Software 95(13) 1-33
  7. Overstall, A.M., Woods, D.C. & Parker, B.M. (2020). Bayesian optimal design for ordinary differential equation models with application in biological science. Journal of the American Statistical Association 115 583-598
  8. Overstall, A.M., & McGree, J.M. (2020). Bayesian design of experiments for intractable likelihood models using coupled auxiliary models and multivariate emulation. Bayesian Analysis 15(1) 103-131
  9. Overstall, A.M., Woods, D.C. & Martin, K.J. (2019). Bayesian prediction for physical models with application to the optimization of the synthesis of pharmaceutical products using chemical kinetics. Computational Statistics and Data Analysis 132 126-142.
  10. Heck, D.W., Overstall, A.M., Gronau, Q.F. & Wagenmakers, E. (2019). Quantifying Uncertainty in Transdimensional Markov Chain Monte Carlo Using Discrete Markov Models. Statistics and Computing 29 (4) 631-643
  11. Overstall, A.M., McGree, J.M. & Drovandi, C.C. (2018). An approach for finding fully Bayesian optimal designs using normal-based approximations to loss functions. Statistics and Computing 28 (2) 343-358
  12. Overstall, A.M. & Woods, D.C. (2017). Bayesian Design of Experiments using Approximate Coordinate Exchange. Technometrics 59 458-470
  13. Woods, D.C., Overstall, A.M., Adamou, M., & Waite, T.W. (2017). Bayesian design of experiments for generalised linear models and dimensional analysis with industrial and scientific application (Invited paper with Discussion) Quality Engineering 29(1) 91-103
  14. Overstall, A.M. & Woods, D.C. (2016). Multivariate emulation of computer simulators: model selection and diagnostics with application to a humanitarian relief model. Journal of the Royal Statistical Society Series C 65(4) 483-505
  15. King, R. & Overstall, A.M. (2015). Population Size Estimation and Capture-Recapture Methods. International Encyclopedia of the Social & Behavioral Sciences (Second Edition). 603-608.
  16. Overstall, A.M. & King, R. (2014). conting: an R package for Bayesian analysis of complete and incomplete contingency tables. Journal of Statistical Software. 58(7) 1-27.
  17. Overstall, A.M., King, R., Bird, S.M., Hutchinson, S.M. & Hay, G. (2014). Incomplete contingency tables with censored cells with application to estimating the number of people who inject drugs in Scotland. Statistics in Medicine. 33(9) 1564-1579.
  18. King, R., Bird, S.M., Overstall, A.M., Hay, G. & Hutchinson, S.M. (2014). Estimating prevalence of injecting drug users and associated heroin-related death rates in England by using regional data and incorporating prior information. Journal of the Royal Statistical Society Series A. 177(1) 209-236.
  19. Overstall, A.M., & King, R. (2014). A default prior distribution for contingency tables with dependent factor levels. Statistical Methodology. 16(1) 90-99.
  20. King, R., Bird, S.M., Overstall, A.M., Hay, G. & Hutchinson, S.M. (2013). Injecting drug users in Scotland, 2006: listing, number, demography, and opiate-related death-rates. Addiction Research and Theory. 21(3) 235-246.
  21. Overstall, A.M. & Woods, D.C. (2013). A Strategy for Bayesian Inference for Computationally Expensive Models with Application to the Estimation of Stem Cell Properties. Biometrics. 69(2) 458-468.
  22. Forster, J.J., Gill, R.C. & Overstall, A.M. (2012). Reversible jump methods for generalised linear models and generalised linear mixed models. Statistics and Computing. 22(1) 107-120.
  23. Overstall, A.M. & Forster, J.J. (2010). Default Bayesian model determination methods for generalised linear mixed models. Computational Statistics and Data Analysis. 54(12) 3269-3288.