Functional Panel Data Analysis: Harnessing Big Data for Society

Panel data models measure cross-sections of complex data sets, such as subnational demographic and high-frequency financial data. They are critical for solving pressing societal and economic challenges, yet current methods fail to capitalise on opportunities afforded by unprecedented volumes and sources of data. This project aims to develop new theories, better methods and faster algorithms to analyse function-valued panel data to produce accurate forecasts. Expected outcomes involve advances in big data-analytic theory and easy-to-implement algorithms for forecasting demographic and financial data. Improved forecasts will bring significant benefits for financial markets, superannuation and insurance industries, governments and society.