Skip to Content
Bayesian statistics offer a powerful paradigm for the development of scientific knowledge. Our research includes probabilistic modelling, stochastic computation, the use and implementation of Monte Carlo Markov Chains, and other approaches in Bayesian inference and decision theory.
We work on the development of computationally efficient statistical methods for complex models with applications in genetics, and MCMC algorithms with applications in epidemiology, geophysics and transportation.
The success of businesses, government and institutions is increasingly dependent on their ability to transform data into information, insights and novel data-products. We have expertise in data analysis in a number of industry sectors to work with large datasets, both stored and incoming to discover knowledge in the mountains of data.
We apply cutting-edge methods from computer science and statistics to model patterns and relationships, typically using large datasets. We research and apply techniques such as machine learning, image analysis, and pattern recognition to model data from a wide variety of fields including business analytics, social media analysis, genetics and ecology.
The plethora of data available in large systems (such as genetics, ecology, bioinformatics and the health sciences) requires the analysis of multiple variables simultaneously. We research and develop rigorous statistical methods for the analysis and visualisation of high-dimensional systems, graphical models, ordination, and spatial statistics.
We have strengths in the development of statistical methodology and accompanying theory for smoothing methods; spatial statistics; reliability and survival analysis; and network tomography.
We create statistical tools for ecological applications, including models of ecological systems, species’ abundances, biodiversity and community ecology. We engage in active field-based ecological research, and consulting for environmental monitoring and impact assessment.
Our scientists are developing new methods to assess quality in manufacturing processes. We have particular expertise in applications in the dairy industry.
We have expertise in statistical modelling and inference for transportation systems. Particular emphases include the development and analysis of models to describe the day-to-day dynamics of traffic flows on road networks, and methods of inference for traffic models using low-dimensional data (network tomography).
We work on statistical problems in geophysics. Our focus is on the spatio-temporal estimation of hazard, especially from volcanoes or earthquakes, and the assessment of the risk.
Find programmes with a research element, including the PhD.
Search our staff database for an expert or area of expertise.
Statistical inverse problems occur when we wish to learn about some phenomenon that is observed only indirectly, or with error. For example in ecology, where we want to count the number of times individual animals are (re)sighted but identification errors may occur.
A $535,000 grant from the 2017 Marsden Fund is allowing Martin Hazelton and a team of statisticians from Massey to look at techniques from algebraic statistics to develop and study new polytope samplers that can do the above effectively and efficiently.
Professor Geoff Jones was awarded the Littlejohn Research Award by the New Zealand Statistical Association in 2016. The Award is the Association's senior research honour. It recognises original statistical research published in the last five calendar years.Professor Geoff Jones - 2016
Littlejohn Award for Massey statistician
Professor Martin Hazelton was awarded $535,000 from the Royal Society's Marsden Fund for the research project 'Lattice polytope samplers: theory, methods and applications'.Professor Martin Hazelton - 2017
Marsden funding for lattice polytope sampler research