## Research

Below are short descriptions of some of our research projects and themes (past and present).

- Statistics in Manufacturing (more)
- Radiocarbon calibration (more)
- Heaton T. J., Blackwell P. G., Buck C. E. (2009). A Bayesian approach to the estimation of radiocarbon calibration curves: the Intcal09 methodology. Radiocarbon, 51, 1151-1164.
- Blackwell P.G. and Buck C.E. (2008). Estimating radiocarbon calibration curves. Bayesian Analysis, 3, 225-248. With discussion; rejoinder by PGB and CEB, pp. 263-268.
- BCal: an on-line Bayesian radiocarbon calibration tool.
- Statistical Ecology (more)
- Harris, K. J. and Blackwell, P. G. Flexible continuous-time modelling for heterogeneous animal movement. Ecological Modelling, 255, 29-37.
- Genetics (more)
- Morrissey, E.R., Juárez, M.A., Denby, K.J. and Burroughs, N.J. (2011). Inferring the time-invariant topology of a non-linear sparse gene regulatory network using fully Bayesian spline autoregression. Biostatistics, 12, 682-694 (DOI:10.1093/biostatistics/kxr009).
- Morrissey, E.R., Juárez, M.A., Denby, K.J. and Burroughs, N.J. (2010). On reverse engineering of gene interaction networks using time course data with repeated measurements. Bioinformatics, 26, 2305-2312 (DOI:10.1093/bioinformatics/btq421).
- Bayesian Statistics in Health Economics (more)
- Strong, M. and Oakley, J. E. (2013). An efficient method for computing single parameter partial expected value of perfect information. Medical Decision Making, 33, 755-766.
- Strong, M., Oakley J. E. and Chilcott, J. (2012). Managing structural uncertainty in health economic decision models: a discrepancy approach. Journal of the Royal Statistical Society, Series C, 61(1), 25-45.
- Bayesian Statistics in Finance (more)
- Triantafyllopoulos, K. (2012). Multivariate stochastic volatility modelling using Wishart autoregressive processes. Journal of Time Series Analysis, 2012, 33, 48-60.
- Juárez, M. A. and Steel, M. F. J. (2010). Model-based Clustering of non-Gaussian Panel Data based on skew-t distributions. Journal of Business and Economic Statistics, 28, 52-66 (DOI: 10.1198/jbes.2009.07145).
- Juárez, M. A. and Steel, M. F. J. (2010). Non-Gaussian Dynamic Bayesian Modelling for Panel Data. Journal of Applied Econometrics, 25, 1128-1154 (DOI: 10.1002/jae.1113).
- Eliciting prior probability distributions (more)
- Gosling, J.P., Oakley, J.E. and O'Hagan, A. (2007). Nonparametric elicitation for heavy-tailed prior distributions. Bayesian Analysis, 2, 693-718.
- O' Hagan, A., Buck, C. E., Daneshkhah, A., Eiser, J. E., Garthwaite, P. H., Jenkinson, D. J., Oakley, J. E. and Rakow, T. (2006). Uncertain Judgements: Eliciting Expert Probabilities. Chichester: Wiley.
- The Sheffield Elicitation Framework (SHELF)
- The MATCH Uncertainty Elicitation Tool
- Uncertainty quantification for complex computer models (more)
- Genetic information in preventative medicine (more)
- Studies in autism (more)
- Bayesian clinical trial simulation (more)
- Nixon, R.M., O'Hagan, A., Oakley, J. E., Madan, J., Stevens, J.W. Bansback, N. and Brennan, A. (2009). The Rheumatoid Arthritis Drug Development Model: A case study in Bayesian clinical trial simulation. Pharmaceutical Statistics 8(4), 371-389
- Ren, S. and Oakley, J. E. (2013). Assurance calculations for planning clinical trials with time-to-event outcomes. To appear in Statistics in Medicine.

We have developed Bayesian statistical models to construct internationally agreed radiocarbon calibration curves, in a project funded by NERC. We also coordinate the development of on-line Bayesian radiocarbon calibration software known as BCal, which we host.

Sheffield is one of the partner institutions in the EPSRC and NERC funded National Centre for Statistical Ecology. This has driven research on inference for switching multivariate diffusion processes, motivated by applications to wildlife telemetry.

We are working closely with colleagues around the University to develop statistical methods to use data arising from high throughput genotyping technologies (arrays and next generation sequencing) to identify DNA variants influencing disease risk. We aim to devise methods to fully exploit the increasingly detailed genomic information becoming available.

Other research includes developing probabilistic models for, and making inferences about, sequence level DNA methylation processes. We have developed graphical models and Bayesian regulatory networks for inference of gene and microRNA interactions, which can be used for targeting specific diseases.

We have a long-running collaboration with researchers in the School of Health and Related Research on the application of Bayesian Statistics in Health Economics. Current interests include experimental design for discrete choice experiments in health state utility surveys, and methods for quantifying uncertainty in health economic model predictions due to uncertainty in the model structure.

We have developed Bayesian models for financial time series. Our work involves Bayesian analysis of statistical arbitrage, volatility and trading models. We have also developed hierarchical Bayesian statistical methods for classification (clustering) of panel (longitudinal) data in economics. We have also developed methods for Bayesian cointegration for financial time series and panel economic data.

Within the field of Bayesian statistics, we are particularly interested in how to elicit proper prior distributions. Members of the group have co-authored a textbook, and have written a protocol and software for elicitation: The Sheffield Elicitation Framework (SHELF). We have also collaborated in producing a web-based version of the SHELF software: The MATCH Uncertainty Elicitation Tool.

Sheffield was the lead institution in a consortium of five universities working on the Managing Uncertainty in Complex Models project (MUCM), funded by RCUK. The MUCM project was concerned with issues of uncertainty in simulation models (also known as process models, mechanistic models, computer models, etc.). MUCM techniques address questions such as uncertainty quantification, uncertainty propagation, uncertainty analysis, sensitivity analysis, calibration (or tuning, history matching, etc.) and ensemble analysis.

We have investigated the use of prior information and simulation in clinical trial design. The aim is to assess the probability that a clinical trial will produce a successful outcome, which may differ considerably from the the statistical power of the trial. This involves exploiting prior knowledge about the efficacy of the treatment in question.