Efficient Bayesian analysis by exploiting fast approximate models. This project will develop novel methods for efficient statistical analysis of complex stochastic models. It will generate new knowled

Description

Efficient Bayesian analysis by exploiting fast approximate models. This project will develop novel methods for efficient statistical analysis of complex stochastic models. It will generate new knowledge in the area of statistics by exploiting and correcting fast model approximations to accelerate parameter estimation. The expected outcome of the project is the ability to process a much wider class of statistical models that are very computationally intensive to simulate on a computer. The methods are motivated by important problems in systems biology and epidemiology, but will be widely applicable, enabling discoveries in many fields. Other benefits include reduced reliance on supercomputing resources and training of the next generation of statistical data scientists.. Scheme: Discovery Projects. Field: 4905 - Statistics. Lead: Prof Christopher Drovandi