Event

Individualised treatment, which relies on the ability to identify and prescribe subject-specific treatments, is an important application of using data in the context of decision-making.

For this purpose, it is crucial to identify the variables that impact the optimal treatment decision. Typical variable selection techniques focus on a different target, in that they select variables that are important for prediction, which are not necessarily those that are important for treatment assignment. In fact, the two sets of variables are often quite different in practice. We propose a general framework to test for, and identify, the important variables for the goal of treatment decision-making.

Howard Bondell is Professor of Statistical Data Science in the School of Mathematics and Statistics at the University of Melbourne, where he has been since 2018. Professor Bondell is a current ARC Future Fellow (2020-2024), and Co-Director of the Melbourne Centre for Data Science.

Professor Bondell received his Ph.D. in Statistics from Rutgers University in 2005 and immediately commenced his academic career in the Department of Statistics at North Carolina State University. He was elected Fellow of the American Statistical Association in 2017. His current research interests include: model selection, robust estimation, regularisation, Bayesian methods, and all aspects of modelling and handling uncertainty in statistical and machine learning approaches.