Talk details: Causal identification is a challenging but fundamental task in many fields, including when studying complex dynamical systems wherein experiments or model-based regressions may not be available or appropriate. Dealing with the complexity of real-world phenomena requires tools that can characterise and test causality in nonlinear dynamical systems. One promising method is called empirical dynamic modelling (EDM), which allows distinguishing causation from mere correlation while making minimal assumptions about nonlinearity, stability, and equilibrium of the underlying complex system. This presentation introduces the key concepts of the empirical dynamic modelling and discusses some examples using the edm package in Stata.
Additional Information
Speaker: Dr Jinjing Li is a Professor at the National Centre for Social and Economic Modelling (NATSEM), University of Canberra. He is an internationally recognised expert in microsimulation modelling and serves as a board member of the International Microsimulation Association (IMA). His simulation models have been used by academics, policymakers, advocacy groups as well as independent organisations. In 2018, Prof Li was invited to give evidence at a Senate Committee Inquiry into the Treasury Laws Amendment due to his expertise in tax simulation modelling.