Maximum penalised likelihood MPL for the Cox model
New computational methods for MPL estimation and inference in the Cox model and its extensions are developed and studied in this project. This approach allows greater modelling flexibility while being superior in terms of statistical accuracy and bias.
This feature is even neater when high levels of censoring are present in the data (eg. ADVANCE, SAFE, DRIVE and mTBI studies). Collaboration with Associate Professor J. Ma, Macquarie University, who is leading the project.