Skip to content


Statistics plays a vital role in the development and practice of modern science, and is vigorous area of research within the mathematical sciences. It is also an essential element of modern data science.

Research in Statistics at the University of Adelaide covers a broard spectrum of applications in areas such as medical research, biology and hydrology in collaboration with other leading researchers, as well as the development of new statistical theory and methods needed to deal with modern data problems.

Researcher and Interests
Gary Glonek Design and analysis of microarray and other experiments, data mining, statistical computing, Bayesian analysis, analysis of categorical data, biostatistics, gene expression studies, epidemiology
Andrew Metcalfe Applications of Statistics, in particular to problems in hydrology and water resources. The analyses typically requires a multivariate time series approach or a spatial statistical analysis
Adam (Ben) Rohrlach Phylogenetics, population genetics, Bayesian analysis, biostatistics
Patty Solomon Gene expression studies, design and analysis of microarray and other experiments, analysis of proteomic spectra, gene and protein networks, data mining, components of variance, biostatistics, survival analysis, clinical trials, critical care medicine, monitoring and assessing health outcomes, epidemiology
Simon Tuke Analysis of microarray data, time course microarray experiments, biostatistics, statistical equivalence

Research Areas




Applied Statistics

Possible PhD Projects

  • Modelling large networks - Dr Jonathan Tuke
    Interdisciplinary Collaboration

    Networks are ubiquitous - if you want to model the connections between subjects then a network is required. From molecular biology to twitter, there is a strong demand for statistical methods to cope with the influx of data.

    The standard statistical models for networks are exponential random graph models (ERGMs). ERGMs are an excellent starting point, but assume that one model is adequate to cope with the entirety of an observed network. This assumption is dangerous. In this Ph.D., we will investigate the problem of assuming a single model, and then develop methods - a mosaic of networks - that will use random effects modelling to allow heterogeneity in the model.

    Network of interdisciplinary collaboration in ARC grant applications. Each node is a FOR code. Each edge indicates a grant that has both nodes as an FOR code. COlour of nodes indicates the main discipline area.

  • Predicting Australia’s climate with statistical machine learning - Professor Patty Solomon

    Predicting Australia’s climate with statistical machine learning

    The Bureau of Meteorology and CSIRO currently use numerical modelling for weather forecasting and predicting long-term trends in Australia’s climate. This research project will develop a statistical machine learning approach to predicting Australia’s climate with the intention of establishing the utility of deep learning as an alternative to current climate models. Temporal and spatial data records are available on air and sea temperatures, rainfall and other atmospheric variables, sometimes dating back more than 100 years. Short- and long-term climate forecasts will be compared, and a series of case-studies on snowfall, wheat and grape production will be developed.

School of Mathematical Sciences
Level 6, Ingkarni Wardli

North Terrace Campus
See location on map


General email
Head of School email
Telephone: +61 8 8313 5407
Facsimile: +61 8 8313 3696

Find Us Online

School of Maths on Facebook    School of Maths Blog