Adaptive treatment and robust control

This studentship is available as part of a three-way collaboration between researchers in statistics, in engineering, and in mathematical analysis. The aim is to solve an array of problems in the general area of adaptive intervention, through the integration of techniques and approaches that have been developed in distinct ways in the three fields, specifically optimal dynamic treatment methodology in statistics and control theory in engineering and analysis. The studentship is to be primarily focused on the statistical aspects of this programme and the general statistical areas include causal inference, longitudinal data analysis and missing data problems.

The specific task is to derive adaptive decision rules for medical treatments or other interventions based on subject-specific characteristics and individual biomarker trajectories. For example, the optimum dose of warfarin anticoagulation treatment varies over time as a function of lifestyle and other patient-specific factors. The aim is to build a decision rule which uses the patient’s prior history to provide clinicians with guidance as to the best dose to prescribe. Other applications include chemotherapy for leukaemia patients, controlling animal growth rate and training of racing horses.