Statistical Inference for Probability-Linked Longitudinal Data. The Strategic Roadmap for the Australian Government's National Collaborative Research Infrastructure Strategy states that analysis of linked data, and particularly linked longitudinal data, has the potential to revolutionise Australian public health research. Similar benefits should flow from analysis of linked datasets in other areas, e.g. the Statistical Longitudinal Census Dataset that the Australian Bureau of Statistics intends ....Statistical Inference for Probability-Linked Longitudinal Data. The Strategic Roadmap for the Australian Government's National Collaborative Research Infrastructure Strategy states that analysis of linked data, and particularly linked longitudinal data, has the potential to revolutionise Australian public health research. Similar benefits should flow from analysis of linked datasets in other areas, e.g. the Statistical Longitudinal Census Dataset that the Australian Bureau of Statistics intends to create by linking individual records across censuses. These benefits will be maximised by controlling the impact of linkage error when analysing these datasets. This proposal will develop the statistical theory and related methodology to solve this problem in a statistically efficient manner.Read moreRead less
Australian Laureate Fellowships - Grant ID: FL200100176
Funder
Australian Research Council
Funding Amount
$3,128,080.00
Summary
Theoretical Foundations of Ethical Machine Learning. The project will develop a systematic theory of ethical machine learning. Machine learning is a powerful and pervasive technology that is already having a huge impact on Australia. When applied to data about people there are a range of ethical harms that can arise (fairness, and privacy are two of them). The project will develop a rigorously grounded foundation for managing such ethical harms. For example it will allow the quantification of t ....Theoretical Foundations of Ethical Machine Learning. The project will develop a systematic theory of ethical machine learning. Machine learning is a powerful and pervasive technology that is already having a huge impact on Australia. When applied to data about people there are a range of ethical harms that can arise (fairness, and privacy are two of them). The project will develop a rigorously grounded foundation for managing such ethical harms. For example it will allow the quantification of the inevitable trade-offs between fairness and utility. The benefits of the project will include the best possible ways of managing these trade-offs, competitive advantage for Australian firms developing the technology, and will ensure that the country retains a social license to use the technology.Read moreRead less
New methods for modelling real-world extremes. This project aims to develop new theory and methods for analysing and predicting extreme values observed in real-world processes. Many existing techniques are limited by convenient mathematical assumptions that commonly do not hold in practice: dependence at asymptotic levels, process stationarity, and that the observed data are direct measurements of the process of interest. As a result, using these techniques may produce undesirable results. Expec ....New methods for modelling real-world extremes. This project aims to develop new theory and methods for analysing and predicting extreme values observed in real-world processes. Many existing techniques are limited by convenient mathematical assumptions that commonly do not hold in practice: dependence at asymptotic levels, process stationarity, and that the observed data are direct measurements of the process of interest. As a result, using these techniques may produce undesirable results. Expected outcomes of this project include theoretically justified data analysis techniques that can accurately model extreme values seen in the real world. Project benefits include more realistic analyses of nationally important applications in climate, bushfire insurance risk, and anomaly detection.Read moreRead less
A likelihood-based approach to combined surveys inference. This project focuses on the development of statistical theory for efficient integration of information across multiple complex sample surveys. It will develop theory and methodology that will answer complex questions about relationships between important social, economic and health related variables that are presently measured in separate surveys.
Asymptotic Expansions and Large Deviations in Probability and Statistics: Theory and Applications. Statistics is the major enabling science in a number of disciplines. This is fundamental research in probability and statistics but it has wide applications in Biology and Social Sciences which will ultimately be of national benefit. The behaviour of self normalized sums is an exciting new area of fundamental research that has implications for the application of statistics in many areas. U-statist ....Asymptotic Expansions and Large Deviations in Probability and Statistics: Theory and Applications. Statistics is the major enabling science in a number of disciplines. This is fundamental research in probability and statistics but it has wide applications in Biology and Social Sciences which will ultimately be of national benefit. The behaviour of self normalized sums is an exciting new area of fundamental research that has implications for the application of statistics in many areas. U-statistics for dependent situations has direct application to understanding financial time series and the analysis of sample survey data. Saddlepoint methods provide extremely accurate approximations in a number of important applications.
Read moreRead less
Empirical saddlepoint approximations and self-normalized limit theorems. Finite population sampling and resampling methods such as the bootstrap and randomization methods are central in a number of areas of application and M-estimates are the major method used to give robust methods under mild conditions; in both these areas statistics are used which are Studentized or self-normalized. We will develop asymptotic approaches for such statistics. Saddlepoint and empirical saddlepoint methods will ....Empirical saddlepoint approximations and self-normalized limit theorems. Finite population sampling and resampling methods such as the bootstrap and randomization methods are central in a number of areas of application and M-estimates are the major method used to give robust methods under mild conditions; in both these areas statistics are used which are Studentized or self-normalized. We will develop asymptotic approaches for such statistics. Saddlepoint and empirical saddlepoint methods will be used to give methods which have second order relative accuracy in large deviation regions and we will obtain limit results and Edgeworth approximations. Emphasis will be on obtaining results under weak conditions necessary for applications.Read moreRead less
Rare Event Simulation with Heavy Tails. The project provides a rigorous way to enhance our understanding of the mechanisms that bring about catastrophic rare events such as urban flooding, electricity shortages and financial bankrupcy. Australia is at the forefront of exciting recent developments in rare event simulation. The advancement of the knowledge in this area will generate a competitive advantage for various sections of the Australian industry, including the areas of industrial reliabili ....Rare Event Simulation with Heavy Tails. The project provides a rigorous way to enhance our understanding of the mechanisms that bring about catastrophic rare events such as urban flooding, electricity shortages and financial bankrupcy. Australia is at the forefront of exciting recent developments in rare event simulation. The advancement of the knowledge in this area will generate a competitive advantage for various sections of the Australian industry, including the areas of industrial reliability, finance and insurance, were accurate simulation techniques are becoming increasingly important.Read moreRead less
Fundamental Studies in System Identification. To operate a dynamic system such as a chemical process plant or an economy one needs two things; the equations describing the system; a way of regulating the system to provide desired outcomes. System identification provides the first; control engineering design provides the second. This proposal addresses three important problems in system identification and control. Firstly since the equations can never be known precisely we aim to determine what i ....Fundamental Studies in System Identification. To operate a dynamic system such as a chemical process plant or an economy one needs two things; the equations describing the system; a way of regulating the system to provide desired outcomes. System identification provides the first; control engineering design provides the second. This proposal addresses three important problems in system identification and control. Firstly since the equations can never be known precisely we aim to determine what is the best one can do? Secondly to provide then tight error bounds for the control design;
thirdly to develop new methods for some hitherto unresolved problems in system identification.Read moreRead less
Stein's method for probability approximation. Data of counts in time, such as incoming calls in telecommunications and the clusters of palindromes in a family of herpes-virus genomes, arise in an extraordinarily diverse range of fields from science to business. These problems can be modelled by sums of random variables taking values 0 and 1 in probability theory, thus permitting approximate calculations which are often good enough in practice. This project will obtain such approximate solutions ....Stein's method for probability approximation. Data of counts in time, such as incoming calls in telecommunications and the clusters of palindromes in a family of herpes-virus genomes, arise in an extraordinarily diverse range of fields from science to business. These problems can be modelled by sums of random variables taking values 0 and 1 in probability theory, thus permitting approximate calculations which are often good enough in practice. This project will obtain such approximate solutions and estimate the errors involved. Applications include analysis of data in insurance, finance, flood prediction in hydrology.Read moreRead less
Discovery Early Career Researcher Award - Grant ID: DE200100435
Funder
Australian Research Council
Funding Amount
$365,039.00
Summary
Modern statistical methods for complex multivariate longitudinal data. The project aims to develop novel approaches for the statistical analysis of large, complex multivariate longitudinal data, and apply them to two datasets to address scientific questions related to the drivers and consequences of poor physical and mental health in Australia, and the spatio-temporal evolution of species assemblages in the Southern Ocean. The project expects to develop new knowledge in the areas of statistical ....Modern statistical methods for complex multivariate longitudinal data. The project aims to develop novel approaches for the statistical analysis of large, complex multivariate longitudinal data, and apply them to two datasets to address scientific questions related to the drivers and consequences of poor physical and mental health in Australia, and the spatio-temporal evolution of species assemblages in the Southern Ocean. The project expects to develop new knowledge in the areas of statistical model building, model selection, and inference for multivariate longitudinal data. This will lead to a suite of modern methods and insights for computationally efficient, mathematically rigorous statistical data analysis that, when applied, should provide significant benefits to public health and ecology.Read moreRead less