Efficient Estimation of Statistical Models with Many Parameters. Statistical models are used extensively in business, engineering and the sciences to describe the behavior of systems subject to uncertainty. There are often many unknowns in such models and relatively little data to estimate them. The object of the research is to develop methods that make these statistical models practical to use. The research team will apply the methodology to solve problems in economics, finance, marketing and t ....Efficient Estimation of Statistical Models with Many Parameters. Statistical models are used extensively in business, engineering and the sciences to describe the behavior of systems subject to uncertainty. There are often many unknowns in such models and relatively little data to estimate them. The object of the research is to develop methods that make these statistical models practical to use. The research team will apply the methodology to solve problems in economics, finance, marketing and the analysis of gene expression data. The project will also train doctoral and postdoctoral students and enhance Australia's reputation for research excellence in the Statistical and Mathematical Sciences. Read moreRead less
Bayesian estimation of flexible spatial models with applications in medical imaging and econometric modeling. This project aims to develop statistical methodology for estimating flexible highly parameterised Bayesian spatial models. The flexible models examined will include regression, choice and time series models for data that is spatially registered. Spatial smoothing of parameters in the models will involve application of hierarchical spatial prior distributions. The resulting methodology wi ....Bayesian estimation of flexible spatial models with applications in medical imaging and econometric modeling. This project aims to develop statistical methodology for estimating flexible highly parameterised Bayesian spatial models. The flexible models examined will include regression, choice and time series models for data that is spatially registered. Spatial smoothing of parameters in the models will involve application of hierarchical spatial prior distributions. The resulting methodology will be applied to the analysis of medical imaging data and to the estimation of spatial econometric models of residential real estate prices. The expected outcomes include developments in the frontier framework of Bayesian computational estimation methodology, improved methods for medical image processing and estimation of high resolution spatial models of residential real estate prices in Australian metropolitan centres.Read moreRead less
Prior sensitivity analysis for Bayesian Markov chain Monte Carlo output. This project aims to develop the first set of techniques to implement an automated output sensitivity analysis for Markov Chain Monte Carlo (MCMC) estimation methods. Computationally intense Bayesian MCMC provide a powerful alternative to classical methods for the estimation of economic models. An obstacle to their wider application is that researchers need to specify prior beliefs about model parameters that will affect t ....Prior sensitivity analysis for Bayesian Markov chain Monte Carlo output. This project aims to develop the first set of techniques to implement an automated output sensitivity analysis for Markov Chain Monte Carlo (MCMC) estimation methods. Computationally intense Bayesian MCMC provide a powerful alternative to classical methods for the estimation of economic models. An obstacle to their wider application is that researchers need to specify prior beliefs about model parameters that will affect the results. The expected outcomes will enable researchers to undertake a routine assessment of the sensitivity of the results to prior inputs.Read moreRead less
Estimation and Inference in Weakly Identified Models. Economic and social systems are made up of interacting components leading to complex structures that are difficult to predict and manage. Consequently policy analysis and decision-making must be informed by statistical analysis of data. In many situations the informational content of observations is minimal; examples of such situations are found in the areas of education, health, finance and various aspects of macroeconomic analysis. This pro ....Estimation and Inference in Weakly Identified Models. Economic and social systems are made up of interacting components leading to complex structures that are difficult to predict and manage. Consequently policy analysis and decision-making must be informed by statistical analysis of data. In many situations the informational content of observations is minimal; examples of such situations are found in the areas of education, health, finance and various aspects of macroeconomic analysis. This project aims to develop methods of estimation and inference that make more efficient use of the information available in data. This will lead to more precise statistical analyses, resulting in a clearer understanding of economic and social systems, and better informed policy analysis and decision-making.Read moreRead less
New approaches for testing in nonlinear models. The outcome of this project is a new econometric methodology that will be particularly useful for developing our understanding of Australian (and global) financial markets. Specific benefits are that (i) our value-at-risk models will enhance national and international awareness of issues relating to financial risk management; (ii) our exchange rate pass through model will aid the development of Australian trade and pricing policies and (iii) our du ....New approaches for testing in nonlinear models. The outcome of this project is a new econometric methodology that will be particularly useful for developing our understanding of Australian (and global) financial markets. Specific benefits are that (i) our value-at-risk models will enhance national and international awareness of issues relating to financial risk management; (ii) our exchange rate pass through model will aid the development of Australian trade and pricing policies and (iii) our duration models for trade in Australian stocks will lead to a better understanding of the microstructure of the Australian stock market.Read moreRead less
Asymptotics in non-linear cointegrating regression: theory and applications. This project provides fundamental research in statistics, econometrics and probability. The results on martingales and nonlinear functionals of integrated stochastic processes will apply to a range of statistical, empirical finance and economic models.
Non-linear cointegrating regression with endogeneity. This project aims to develop the asymptotic theory of estimation and statistical inference in models concerned with non-linear co-integrating regression with endogeneity and long memory. This project will tackle a number of long-standing technical problems related to non-linear covariance functionals and non-linear transformation of nonstationary time series. This project is intended to provide technical tools for practitioners to study the l ....Non-linear cointegrating regression with endogeneity. This project aims to develop the asymptotic theory of estimation and statistical inference in models concerned with non-linear co-integrating regression with endogeneity and long memory. This project will tackle a number of long-standing technical problems related to non-linear covariance functionals and non-linear transformation of nonstationary time series. This project is intended to provide technical tools for practitioners to study the long-run relationship of economic variables, and could apply to a range of statistical, empirical finance and economic models, enhancing national leadership in these areas.Read moreRead less
Development of general methodology for estimating complex time series models. This project will develop novel methods and models for analysing socio-economic and financial data measured over time and will illustrate them with applications. The methods will allow for more efficient and more accurate processing of information and better forecasting which will facilitate better management and more timely policy response.
A new look at modelling population heterogeneity in econometric study. This research will advance existing quantitative techniques in economic study. New theoretical results will help enhance Australian research reputations. The innovative techniques developed in this project will be demonstrated to study labour force participation of people with disabilities in Australia. Findings of the empirical study will help governments in providing financial assistance to affected families and addressing ....A new look at modelling population heterogeneity in econometric study. This research will advance existing quantitative techniques in economic study. New theoretical results will help enhance Australian research reputations. The innovative techniques developed in this project will be demonstrated to study labour force participation of people with disabilities in Australia. Findings of the empirical study will help governments in providing financial assistance to affected families and addressing the issue of labour shortage in Australia. Furthermore the participation of a high profile international researcher will benefit the local research community and provide a research training opportunity for local postgraduate students.Read moreRead less
A Multivariate Dynamic Factor Model of the Australian Business Cycle: Specification, Estimation and Empirical Results. The project aims to extend greatly existing models of national and international business cycles by developing a general class of dynamic factor models for Australia. The project provides a significant contribution to business cycle modelling by solving the intractability problems common to existing classes of dynamic factor models. A key innovation is the development of a simul ....A Multivariate Dynamic Factor Model of the Australian Business Cycle: Specification, Estimation and Empirical Results. The project aims to extend greatly existing models of national and international business cycles by developing a general class of dynamic factor models for Australia. The project provides a significant contribution to business cycle modelling by solving the intractability problems common to existing classes of dynamic factor models. A key innovation is the development of a simulation based estimator to circumvent the statistical and computational problems associated with existing estimators. The expected outcome of the project will be a more reliable way to monitor the phases of the cycle and forecast turning points, which will be of substantial national benefit.Read moreRead less