The COVID-19 pandemic has highlighted the importance on informed and grounded statistical information for policy implementation and decision-making; widespread economic and mental health impacts hang in balance with a need to manage the immediate health risks of a pandemic. Reliable parameter estimation methods for complex stochastic epidemiological models are needed to deliver better decision-making information to governing bodies for future epidemics. The research project hopes to implement indirect inference approaches in approximate Bayesian computation (ABC) on several COVID-19 models to achieve better approximations of the posterior distribution and improve model predictions. The key objectives of this research project are to (1) Identify a set of surrogate models that are well-suited to describing time-series data of pandemics and (2) evaluate and demonstrate the improvements to posterior predictive accuracy and parameter estimates for existing COVID-19 models by using indirect inference approaches for ABC.
Queensland University of Technology
Abhishek is in his final year of Bachelor of Electrical Engineering/Economics degree at QUT. He has both deep and broad talents and interests in data science. He serves as a Research Assistant at QUT and has developed skills through proactive engagement in several research projects with the Australian Research Council Centre of Excellence in Mathematical and Statistical Frontiers (ACEMS). His highlight project has involved estimating and modelling disease spread in a banana plantation which has recently been accepted by PLOS Computational Biology (top two per cent of Modelling and Simulation category according to Scimago).