Tony O’Hagan is the Principal Investigator and leader of the MUCM project. He has been researching in this field for almost 20 years.  His work on uncertainty analysis, building on developments in computer experiments in the 1980s, laid the foundations for the use of Gaussian process emulators for quantifying uncertainty in complex models (with Richard Haylock, 1996/1997).  This was followed by research on codes with different resolutions and calibration (with Marc Kennedy, 2000/2001), funded by an EPSRC Mathematics Programme grant, theory for sensitivity analysis (with Jeremy Oakley, 2004), methods for emuulator validation (with Leo Bastos, 2009) and dynamic emulators (with Stefano Conti and others, 2009/2010).  In 2000, he organised an EPSRC-funded workshop “Statistical analysis of computer code outputs” that for the first time brought together researchers in Gaussian process emulation, computer experiments and other approaches to sensitivity analysis.

Early work featured applications in radiation dosimetry, response to nuclear accidents (with the National Radiological Protection Board and support from the Environment Agency) and sewer system models (supported by WRc).  More substantial applications followed, characterising uncertainty in estimates of carbon fluxes from large vegetation models (with other members of the Centre for Terrestrial Carbon Dynamics, 2006/2008).

More broadly, Tony’s research is in the methodology and applications of Bayesian statistics.  He has developed various other applications of Gaussian processes, including methods for elicitation of expert knowledge that are applicable in formulating uncertainty in model parameters.   Another major strand of his research is Bayesian methods in health econo⍺mics, which also links to the present proposal through the use of complex economic models for evaluating the cost-effectiveness of medicines.  He was influential in the adoption by the National Institute for Health and Clinical Excellence of guidelines for economic evaluation that require probabilistic sensitivity analysis for the explicit recognition of uncertainty.

He is a Professor in the School of Mathematics & Statistics, at the University of Sheffield, which is one of the strongest in the UK for Bayesian statistics and interdisciplinary research.  He has served on the Council of the Royal Statistical Society and the Board of the International Society for Bayesian Analysis, and on the peer review colleges for both the EPSRC and MRC.

Contact Details:
Professor A. O'Hagan
Project Leader, MUCM
School of Mathematics & Statistics
University of Sheffield
Hicks Building
Sheffield
S3 7RH
UK

Phone:            +44 (0) 114-222-3773
Fax:                +44 (0) 114-222-3809
Email:              a.ohagan@sheffield.ac.uk
Web:               http://www.tonyohagan.co.uk/