About us
The increased availability of large data sets and remarkable advances made in computing have been distinguishing features of the 21st century.
Our network links Monash researchers with a select group of internationally renowned econometricians and statisticians from Brown, Cambridge, Columbia, National Tsinghua in Taiwan, Paris-Dauphine, Warwick and Yale Universities, to collaborate on the analysis and application of complex, dynamic and flexible models for high-dimensional statistical data. The aim of the network is to develop new econometric and statistical methods that exploit the power of computing and the information in large data sets, to inform policy decisions related to important global issues such as climate change, energy demand, population ageing and the stability of financial systems.
Addressing such issues requires an understanding of complex dynamic relationships within high-dimensional sets of time series data. A major obstacle in modelling high-dimensional dynamic systems is the curse of dimensionality. Inevitably, one needs to adopt assumptions to help reduce the dimension of the unknown ‘parameters’ of the system. The assumption that a small set of common factors inspired by economic theory drive the dynamics of large systems is widespread, with Ng being a world authority on such factor structures.
Large systems such as climatic and financial systems involve complex trends, and Linton and Phillips are leading authorities on modelling such trends. By teaming up Monash econometricians, such as Anderson, Gao, Hyndman, Poskitt, Vahid and Zhao who are already involved in time series and panel data research programs in the department, with Linton, Ng and Phillips, our network has the propensity to have high impact on the development and use of high-dimensional factor models in practice.
The analysis of high-dimensional dynamic data brings with it many practical challenges.
- how to interactively plot and visualise the data and assess significance of structure discovered by exploratory methods
- how to develop models incorporating rich relationships to estimate the unknown features and use them to produce accurate and informative predictions.
The main aims of the network are to:
- tackle new and challenging issues in modern econometrics and statistics
- use the recognised expertise to establish new research programs in big data analytics
- offer solutions to solve pressing problems in some emerging areas, such as climate, environment, health, ageing, insurance, stability of financial systems, and high-dimensional random matrix theory for deep learning.
We plan to explore all aspects of data analysis including visualisation, modelling, inference and prediction, exploiting the wealth of expertise within the Department itself, and within the group of international experts.
Shmueli is a world-renowned expert in high-dimensional business analytics. She will add her expertise to the existing research program in business analytics led by Cook and Hyndman, in order to tackle the visualisation challenges specifically.
The inferential challenges posed by the dimension and complexity of the envisioned models lend themselves to treatment via statistical methods that draw heavily on simulation and computing power. Robert and Renault are prominent in this area, and together with Martin, Poskitt and others within the department are involved in cutting-edge projects that provide the theoretical foundations for simulation-based techniques, propose new ways of applying such techniques to conduct inference, and use such methods to produce accurate predictions.
Furthering this work will build on well-established expertise and research collaborations to make further high-impact contributions to the statistical analysis of high-dimensional complex systems.
Our network has the potential to lead to significant, long-term, strategic partnerships and benefits.
The establishment of this network will facilitate and enhance collaborative research activities within the department with some of the best and leading centres around the world, such as the Cowles Foundation for Research in Economics at Yale University, Climate Econometrics at the University of Oxford, and the Centre for Research in Econometric Analysis of Time Series at Aarhus University.
The Department of Econometrics and Business Statistics provides a nurturing research environment that is perfectly suited to hosting this network. The network positions Monash Business School well for engaging in broader research agendas and complements the well-established synergies that exist within the boundaries of the business school.
The network is well placed to obtain competitive grant income on several areas addressed by the research clusters. After gaining momentum, there is scope for the network to be upgraded to a University network of excellence and recognition as an ARC Centre of Excellence in Econometrics and Statistics.
The long-term outputs and outcomes of the network’s activities will include work that sheds light on issues that affect society such as the supply of energy, food and health care, as well as publications in leading international journals.
Key international collaborators
- Professor Oliver Linton, Professor of Political Economy, University of Cambridge
- Professor Serena Ng, Professor of Economics, Columbia University
- Professor Peter Phillips, Sterling Professor of Economics and Professor of Statistics, Yale University
- Professor Eric Renault, Professor of Economics, University of Warwick and Brown University
- Professor Christian Robert, Professor of Statistics, Ceremade - Université Paris-Dauphine and University of Warwick (20%)
- Professor Galit Shmueli, Distinguished Professor of Business Analytics, National Tsinghua University, Taiwan.
Principal researchers
- Professor Jiti Gao (Network Director), Professor and Donald Cochrane Chair of Business and Economics
- Professor Heather Anderson, Maureen Brunt Professor of Economics and Econometrics
- Professor Dianne Cook, Professor of Business Analytics
- Professor Rob Hyndman, Professor of Statistics
- Professor Gael Martin, Professor of Econometrics
- Professor Farshid Vahid, Professor of Econometrics.
No content
No content