NIASRA runs a variety of events and seminar series open to academics, students and the community to attend.
Events & seminars
School of Mathematics and Applied Statistics presents:
2021 Statistical Science Lecture
2021 STATISTICAL SCIENCE LECTURER
John C. Malone Professor of Biostatistics and Medicine
The Johns Hopkins University Bloomberg School of Public Health and School of Medicine, Baltimore, USA.
Date: Wednesday 3 November 2021
Time: 10.45am for 11am
Venue: UOW Building 43, Room G01. Also, delivery by Zoom
Machine Learning Taster
The next meeting of the Data and Decision Science Network will be an introduction to machine learning.
When: 12:00 pm - 1:30 pm, 7 October 2021
Where: zoom (link to follow registration)
Register: register at eventbrite
This is a general presentation for anyone curious about what machine learning is (and isn’t).
Today, new technologies are generating measurements for almost everything with the ability to store these massive amounts of variables and data. There is a need to exploit such data; hence, analyses are evolving and are aiming at finding the complex relations in the data. There are many techniques to approach analysis, but some techniques are in much more “fashion” and people are wanting to use them. This is the case of “machine learning” techniques, which are sometimes referred to as if the term “machine learning” was ONE specific technique, and this is observed repeatedly in grants and proposals, where phrases like “we will use machine learning to…” are used, when in fact such a term is an umbrella term for many potential techniques. Given this, it is of importance that researchers get at least a taste of what “machine (statistical) learning” is and what it is not, and what is gained/lost in the use of some of these techniques, plus knowing that it is not a magical “one click” solution. Therefore, the aim of this talk is to give an “in a nutshell” introduction to some of the most commonly used/mentioned techniques in machine learning (ML), while also making the connection to methods in general and statistical techniques. The importance of the trade-offs of variability and bias, while presenting complex concepts in a basic manner to explain what the methods do, and that ML involves more than just “clicking a button”, will be showcased using simple one-on-one variable examples. I will hone on concepts of supervised/unsupervised learning, training and testing data sets, parameters, algorithms and cross-validation.
Presenter: Alberto Nettel-Aguirre, Professor of Biostatistics, Director of the Centre for Health and Social Analytics, NIASRA, UOW
Alberto’s first exposures to ML were in his Postdoctoral years through seminars and application of techniques while mentoring graduate students. He has developed his career working collaboratively in health and medical research and has worked extensively as a biostatistician in a range of projects. His expertise and interests cover biostatistics and methods for health and social research, machine learning, social network analysis and data science, functional data analysis and big graph data. He has incorporated the use of ML techniques (e.g. CART, Nearest Neighbours, Cluster Analysis) in various health research projects in pediatric gastrointestinal disease, neurology, and applications to biomechanics in juvenile arthritis, among others.
|Friday, 24 September, 16:00, Zoom (cancelled)||
Dr Jenny Wadsworth, Lancaster University
Towards higher-dimensional spatial and spatio-temporal extremes
|Friday, 27 August, 14:00, Zoom||
Dr Beverley Gogel, University of Wollongong
|Comparison of tensor product penalised-splines and autoregressive processes for modelling the smooth trend effects in plant breeding field trials|
|Friday, 13 August, 16:30, Zoom||
Dr James Ng, Trinity College Dublin, Ireland
|Mixture of normalizing flows for spherical density estimation|
|Friday, 30 July, 11:30, Zoom||
Professor Robert Deardon, University of Calgary
|Fast parameterization of spatial epidemic models: let’s emulate|
|Friday, 2 July, 11:00, Zoom||
Senior Professor David Steel, University of Wollongong
|Sample design for analysis using high inﬂuence probability sampling|
|Friday, 4 June, 14:00, Zoom||
Dr Luca Maestrini, University of Technology Sydney
|Variational approximations for random effect and latent variable models|
|Friday, 21 May, 14:00, Zoom||
Dr Leah South, Queensland University of Technology
|Monte Carlo variance reduction using Stein operators|
|Friday, 14 May, 16:00, Zoom||
Consulting statistician Jonathan Rougier
|Estimating volume from point-referenced thickness measurements|
|Friday, 30 April, 14:00, Zoom||
Honorary professorial fellow, Siu-Ming, University of Wollongong
|On removing the linkage bias from integrated data sets for analytic inference|
|Friday, 16 April, 14:00, Zoom||
Professor Han Lin Shang, Macquarie University
|Bootstrap rediction bands for functional time series|
|Friday, 9 April, 14:00, Zoom||
Dr David Gunawan, University of Wollongong
|Posterior probabilities for Lorenz and stochastic dominance of Australian income distributions|
|Friday, 19 March, 14:00, Zoom||
Chris Lisle, University of Wollongong
|A new diagnostic to assess information available for variance parameter estimation in Multi-Environment Trial (MET) analyses|
|Friday, 12 March, 14:00, Zoom||
Dr Stephanie Clark, University of Technology Sydney
|Multiple time series analysis with unsupervised and supervised machine learning: Groundwater level patterns in the Namoi region of NSW|
Annual Statistical Science Lecture
in the School of Mathematics and Applied Statistics
- The Statistical Science Lecture (SSL) began in 2018 and is an annual event made possible by a philanthropic donation to the School of Mathematics and Applied Statistics, University of Wollongong.
Statistical Science is the science of uncertainty. More specifically, it is the principled collection, analysis, and interpretation of data, taking into account the uncertainties within and between each of these steps. A critical component of excellent science is the ability to weigh evidence appropriately – statistical thinking lies at the heart of this. The annual Statistical Science Lecture showcases the interdisciplinarity and key role a statistical scientist plays in extracting scientific knowledge from data in the presence of uncertainty.
The inaugural Statistical Science Lecture (SSL) was given in 2018 and is an annual event made possible by a philanthropic donation to the School of Mathematics and Applied Statistics (SMAS), University of Wollongong.
The SMAS 2021 Statistical Science Lecture will be held on Wednesday 3 November 2021
The 2021 Statistical Science Lecturer
Scott L. Zeger
John C. Malone Professor of Biostatistics and Medicine
The Johns Hopkins University Bloomberg School of
Public Health and School of Medicine,
Saving Medical Dollars, Trillions at a Time: A Statistical Perspective
American Medicine wastes $1.2 trillion per year, 5% of GDP, and produces noncompetitive health outcomes. A substantial fraction of this waste is attributable to poor use of information in clinical decisions. This talk addresses a statistical framework that is essential infrastructure to learn from past experience how to improve care for future patients at more affordable costs. We frame four core medical questions in statistical terms, then offer a Bayesian hierarchical modeling approach that integrates complex data with prior biomedical knowledge to address the questions. This approach is starting to be implemented in academic health centers. We use two recent examples from autoimmune disease and Covid-19 clinics to illustrate the methods, identify current obstacles to success, and point out opportunities for data scientists to directly impact future healthcare decisions.
Scott L. Zeger is The Johns Hopkins University’s co-Director of Hopkins in Health, the Johns Hopkins precision medicine partnership of the University, Health System, and Applied Physics Laboratory. He conducts statistical research on regression analysis for correlated responses and on methods for precision medicine. He has made substantive contributions to our understanding of the effects on health of smoking and air pollution, the global etiology of children’s pneumonia, and other topics. Professor Zeger has served as expert witness to the U.S. Department of Justice and several states in their civil suits against the tobacco industry and as a member of the Board of Scientific Advisors for the Merck Research Laboratory. He is a member of the Springer-Verlag editorial board for Statistics and was the founding co-editor of the Oxford University Press journal, Biostatistics. Dr. Zeger’s work has been recognized with several awards including most recently an honorary doctorate from Lancaster University in England and the 2015 Karl Pearson Prize from the International Statistical Institute with Kung-Yee Liang for their development of Generalized Estimating Equations (GEE). Dr. Zeger is most proud of his Golden Apple Awards from the Johns Hopkins Bloomberg School Student Assembly, for excellence in teaching.
The 2020 Statistical Science Lecturer
Professor, University of Sydney, Australia
Sally Cripps is a Professor of Mathematics and Statistics and Director of the ARC Centre in Data Analytics for Resources and Environments (DARE Centre), at the University of Sydney. Sally’s research focus is the development of new and novel probabilistic models which are motivated by the need to solve an applied problem with the potential for impact. She has particular expertise in the use of mixture models for complex phenomena, modelling longitudinal data, nonparametric regression, the spectral analysis of time series, and the construction of transition kernels in MCMC schemes that efficiently explore posterior distributions of interest. Sally is also Chair of the International Society for Bayesian Analysis’ section, Bayesian Education and Research in Practice.
Statistical Science Lecture given on 18 November 2020:
Zen and the Art of Bayesian Geology/Hydrology/Ecology
2020 lecture: Sally Cripps
Noel Cressie, Sally Cripps
Professor Cripps with UOW Students
The 2019 Statistical Science Lecturer
Peter J Diggle
Distinguished Professor, Lancaster University and Health Data Research UK
Peter Diggle is a Distinguished University Professor of Statistics in the Centre for Health Informatics, Computing and Statistics, a teaching and research group within the Lancaster Medical School at Lancaster University working at the interface of statistics, epidemiology, and health informatics. Peter is also Director of Training at Health Data Research UK, working with academic institutions across the UK to draw up a training strategy that builds on existing best practice to create a programme that will transform the careers of future leaders in data science on a national scale. He holds adjunct positions at the Johns Hopkins University, Yale and Columbia Universities, and he was President of the Royal Statistical Society between July 2014 and December 2016.
Statistical Science Lecture given on 06 November 2019:
A Tale of Two Parasites: statistical science to support disease control programmes in Africa
2019 lecture: Peter Diggle
Peter Diggle, V-C Paul Wellings, Noel Cressie
The 2018 Statistical Science Lecturer
Professor, The University of Auckland, New Zealand
Renate Meyer is a Professor in the Department of Statistics at The University of Auckland, with research interests in applied Bayesian inference and MCMC methods. In particular, her research areas comprise time series analysis with applications in astrophysics (gravitational waves), state-space modelling in ecology, multivariate modelling using copulas, survival analysis in medical statistics, and stochastic volatility models for financial time series.
Statistical Science Lecture given on 31 October 2018:
Surfing Gravitational Waves: Black holes and Bayesian nonparametrics
Noel Cressie, Renate Meyer
SMAS morning tea with Renate Meyer
Fellows Research Meetings
The NIASRA Fellows Meetings aim to provide an ongoing opportunity for researchers actively working in areas of interest to the Centre to present both work in progress and recent research results to an audience of fellow researchers and senior professionals who are also working in these areas.Find out more