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
Discovery Early Career Researcher Award - Grant ID: DE150100309
Funder
Australian Research Council
Funding Amount
$328,614.00
Summary
Understanding the Dynamics of Socioeconomic Related Health Inequalities. Health differences across socio-economic groups have persisted in many countries, including Australia, despite decades of considerable improvements in life expectancy and average health status. Little is known of how policies may influence socio-economic health inequalities as the mechanisms underlying them are complex and the causes differ across population groups and over the lifecycle. This project aims to develop method ....Understanding the Dynamics of Socioeconomic Related Health Inequalities. Health differences across socio-economic groups have persisted in many countries, including Australia, despite decades of considerable improvements in life expectancy and average health status. Little is known of how policies may influence socio-economic health inequalities as the mechanisms underlying them are complex and the causes differ across population groups and over the lifecycle. This project aims to develop methods to quantify the major mechanisms that give rise to changes in socio-economic health inequalities in Australia. This project aims to improve our understanding of the dynamic factors that drive changes in health inequalities, thus providing useful information for decision makers about which policies will be cost effective at reducing them.Read moreRead less
Robust methods for heteroscedastic regression models for time series. What is the variability of the exchange rate of the Euro to the Australian dollar? Can the use of the electrocardiogram of a patient be improved as a diagnostic tool for heart disease? A well-known limitation of the existing statistical methods for answering these types of questions is that a small proportion of extreme observations have the potential to lead to results that are more in agreement with the outliers than with bu ....Robust methods for heteroscedastic regression models for time series. What is the variability of the exchange rate of the Euro to the Australian dollar? Can the use of the electrocardiogram of a patient be improved as a diagnostic tool for heart disease? A well-known limitation of the existing statistical methods for answering these types of questions is that a small proportion of extreme observations have the potential to lead to results that are more in agreement with the outliers than with bulk of the data. As a consequence, the statistical analyses may lead to wrong conclusions. This project aims to develop new methodologies to solve this problem for a large class of studies. Applications to stock market risk, exchange rate, and diagnosis of heart diseases will illustrate the new methods.Read moreRead less
Modelling health: Reporting behaviour and misclassification using survey data. Empirical models based on large scale survey data sets are used by health economists to inform policymakers. However, in the case of sensitive topics, a potential for survey misreporting may lead to inaccurate estimates of aberrant behaviours. To date, little work has been done analysing the extent and consequences of inaccurate reporting, especially within health economics. By addressing areas where potential for mis ....Modelling health: Reporting behaviour and misclassification using survey data. Empirical models based on large scale survey data sets are used by health economists to inform policymakers. However, in the case of sensitive topics, a potential for survey misreporting may lead to inaccurate estimates of aberrant behaviours. To date, little work has been done analysing the extent and consequences of inaccurate reporting, especially within health economics. By addressing areas where potential for misinformation is high, the overall quality of results will be enhanced. This research will be submitted to highly ranked health economics and econometrics journals to be made available to relevant policymakers intent on ensuring a healthy society.Read moreRead less
Threshold models in micro-econometrics with applications to empirical models of health. The aim of this project is to develop and apply new statistical approaches to endogenously identify non-linear relationships between explanatory variable(s) and the response variable in non-linear econometric models and to illustrate these with applications important to empirical health economics. Literature proliferates in linear models with non-linear effects, but in health economics non-linear models domin ....Threshold models in micro-econometrics with applications to empirical models of health. The aim of this project is to develop and apply new statistical approaches to endogenously identify non-linear relationships between explanatory variable(s) and the response variable in non-linear econometric models and to illustrate these with applications important to empirical health economics. Literature proliferates in linear models with non-linear effects, but in health economics non-linear models dominate. This project will generalise these techniques to allow for various forms of the threshold variable(s), including categorical and continuous, endogenous and exogenous, and those measured with error.Read moreRead less
Identification Power and Instrument Strength in Discrete Outcome Models. This project aims to develop new econometric and statistical techniques to quantify causal effects in treatment models with discrete outcomes. Expected outcomes include a much-needed weak instrument test, a measure for identification strength in partial identification setting, and an instrument-covariate selection procedure for high dimensional discrete models based identification power. The benefits include advanced knowle ....Identification Power and Instrument Strength in Discrete Outcome Models. This project aims to develop new econometric and statistical techniques to quantify causal effects in treatment models with discrete outcomes. Expected outcomes include a much-needed weak instrument test, a measure for identification strength in partial identification setting, and an instrument-covariate selection procedure for high dimensional discrete models based identification power. The benefits include advanced knowledge in econometrics and statistics, and enhanced tools for program evaluation and policy assessment in empirical causal analysis using observational data. The project falls into the category of smarter information use and is relevant to any national priority areas where policy interventions require assessment.Read moreRead less
Discovery Early Career Researcher Award - Grant ID: DE120101130
Funder
Australian Research Council
Funding Amount
$375,000.00
Summary
New models and estimation methods in nonlinear panel data econometrics. This project will develop new econometric models and methods for capturing dynamic and complex relationships within economic and social systems. The outcomes of this project are expected to improve policy making process concerning climate change, economy and financial markets, through providing accurate estimates of relationships of interest.
Trending time series models with non- and semi-parametric methods. The outcomes of this project will not only complement but also enhance the existing strengths and reputation of Australian researchers in the field of econometrics. The outcomes are also expected to help improve model building and forecasting from better models in climatology, economics, environmetrics and financial econometrics.
Bayesian copula modelling of multivariate dependence: getting to grips with data that is far from normal. Copula models are very popular tools that are changing the way analysts deal with information rich data in fields as diverse as marketing, finance and transport studies. This project aims to improve and extend these tools, so that more accurate and reliable models can be employed, resulting in improved evidence-based decision-making.
Discovery Early Career Researcher Award - Grant ID: DE120101106
Funder
Australian Research Council
Funding Amount
$375,000.00
Summary
Economic rise and decline - as seen from space. This research evaluates the accuracy of night-light based measures of local economic change. Satellite images of night-light cover the entire inhabited regions of the world, thus establishing whether these data can be used to supplement traditional measures of economic activity in countries with weak statistical systems would be a global public good.