"Much of my work in regression modelling deals with data sets that don't conform to linear patterns": An interview with 2013 Hannan Medal Winner Matthew Wand

Features

  • Author: Statistics Views
  • Date: 21 Nov 2013
  • Copyright: Image appears courtesy of Professor Wand

The 2013 Hannan Medal for research in statistical science was awarded earlier this year to Professor Matthew Wand, Distinguished Professor in Statistics in the School of Mathematical Sciences, University of Technology, Sydney, Australia.

The 2013 Hannan Medal, given for research in statistical science by the Australian Academy of Science, was created in honour of the late Professor E J Hannan, FAA, Professor of Statistics at the Research School of Social Sciences of the Australian National University. Professor Hannan made a vast contribution to the world of statistical science with his research into time series analysis.

Wand's main research focus is non-linear statistical models and methodology for high-dimensional and complex data, in the face of rapid technological change. Much of this research incorporates ongoing developments in Machine Learning. His contributions are multifaceted and involve applications, theory, methodology and publicly available software. Whilst most of Wand’s research is generic, areas of application that have driven some his research include public health, computational biology and the natural environment.

Statistics Views talks to Professor Wand about his career and research that led to his award and the growing reputation of Australia and New Zealand in the field of statistics.

thumbnail image:

1. With the teaching career that you are continuing to enjoy, what was it in the first place that inspired you to pursue a career in statistics?

With the research career that I'm continuing to enjoy, what inspired me in the first place to pursue a career in statistics was the attraction of being able to use my mathematical skills to help better analyse data arising from a wide range of areas of application.

2. When and how did you first become aware of statistics as a discipline?

First as a 15-year old in high school, but more seriously as an 18 year-old in my first year of university studies.

3. What are your main areas of interest in statistics?

My current research is driven by data sets becoming bigger and more prevalent with the growth of technology such as the
Internet. Statistics plays a key role in obtaining usfeful information from the massive amounts of data now being generated by science and industry. I'm interested in any application area that can benefit from data analysis. We create generic tools for analysis of datasets, regardless of the particular area of application. The development of these tools involves a combination of statistical modelling, computing and a good deal of `pen and paper' mathematics. Much of my work in regression modelling deals with data sets that don't conform to linear patterns. A current major research direction is modification of this methodology to enable analyses to be done in real time, in response to the increasing prevalence of  streaming data.

4. Your main research focus is non-linear statistical models and methodology for high-dimensional and complex data, in the face of rapid technological change. What is your main objective and what do you hope to achieve through your research?

To make methodology and software for processing data of this type more accessible for such data to be analysed.

5. How far do you think statistics and people's awareness of it has progressed during the last decade in Australia and New Zealand?

Australia and New Zealand are, per capita, two of the strongest countries in terms of use and development of statistics.There
is increasing awareness of data as technology such as web search engines and smart phones become more and more part of people's lives in Australia and New Zealand. The Australian Bureau of Statistics is becoming more sophisticated in its use of recent Statistics and Machine Learning techniques, such as support vector machines, to process census forms. In the Australian business sector business analytics is a major growth area and now many of the `suit and tie' set are taking courses in R.

6. You continue to teach at the School of Mathematical Sciences at the University of Technology, Sydney as Distinguished Professor of Statistics and spent five years as Associate Professor of Biostatistics at Harvard. What do you think the future of statistical research will be?

Future statistical research will need to respond to new types and more voluminous data.

Much of my work in regression modelling deals with data sets that don't conform to linear patterns. A current major research direction is modification of this methodology to enable analyses to be done in real time, in response to the increasing prevalence of streaming data.

7. Do you have any advice for students considering a university  degree in statistics?

The relevancy and usefulness of statistics can only increase in the era of Big Data. My advice is to learn as much about
statistical methodology and computing, and its theoretical underpinning, as you can. This will put you in good stead for
opportunities after you graduate.

8. Over the years, how has your teaching, consulting, and research motivated and influenced each other?

My research and consulting has influenced my teaching and I think statisticians involved in statistics and consulting are better teachers because of it.

9. Do you continue to get research ideas from statistics and incorporate your ideas into your teaching?

Yes. I only do a small amount of teaching these days, but I do continue to incorporate research ideas into my teaching.

10. Where do you get inspiration for your research projects and books?

With research projects, mainly by continually asking: What are the shortcomings of existing statistical methodologies that need fixing for dealing with particular types of data. With books, it is mainly about how to best make a difficult concept more palatable. It is inspirational to know that your books have helped so many people learn such concepts.

11. What has been the most exciting development that you have worked on in statistics during your career so far?

Bringing various sub-fields of statistics together so that statistical analyses can be done in a more unified and efficient manner.

12. What do you think the most important recent developments in the field have been?

One important recent development is approximate Bayesian computing via Monte Carlo and variational methods.

13. What do you think will be the most exciting and productive areas of research in statistics during the next few years?

Statistical methods for massive sets and streaming data. For the latter, the data may be so voluminous that they may not be storable in standard computer memory and therefore need to be processed rapidly on arrival and then discarded.

14. What do you see as the greatest challenges facing the profession of statistics in the coming years?

Training enough young people in the area to meet the demands of current and near-future types of data.

15. Are there people or events that have been influential in your career?

I have been lucky to have excellent mentors throughout my career. An important event was being recruited to the Department of Biostatistics at Harvard University, USA, in 1997. I spent five and a half years there working with medical and public health researchers, and it gave me a much wider perspective of statistics.

Related Topics

Related Publications

Related Content

Site Footer

Address:

This website is provided by John Wiley & Sons Limited, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ (Company No: 00641132, VAT No: 376766987)

Published features on StatisticsViews.com are checked for statistical accuracy by a panel from the European Network for Business and Industrial Statistics (ENBIS)   to whom Wiley and StatisticsViews.com express their gratitude. This panel are: Ron Kenett, David Steinberg, Shirley Coleman, Irena Ograjenšek, Fabrizio Ruggeri, Rainer Göb, Philippe Castagliola, Xavier Tort-Martorell, Bart De Ketelaere, Antonio Pievatolo, Martina Vandebroek, Lance Mitchell, Gilbert Saporta, Helmut Waldl and Stelios Psarakis.