SuSTaIn
Events
Time for Causality workshop
Programme (pdf)
Abstracts (pdf)
Participants (pdf)
Slides for talks (protected)
Photos
Location
Participation
Accommodation
Conference dinner
Local Restaurants
Directions
|
|
Time for Causality - Causal Inference and Dynamic Decisions in Longitudinal Studies
Research workshop: 10-13 April 2012, Bristol, UK
Aims and Objectives
The aim of this workshop is to bring together researchers who are interested in statistical methodology for causal inference and decision making from time-dependent data and modelling of dynamic systems. We envisage that this interest can take various forms. For instance, in many practical applications we may want to evaluate the effect of a strategy for a sequence of decisions, e.g. treatment decisions for patients with a chronic or long-term disease. Moreover, we might be interested in finding a good or optimal decision strategy. Often, the evaluation and optimisation may be hampered by the presence not only of sampling variability but also of time-varying confounding and specialised methods are called for. There may also be applications where possible interventions in the system are less specific and researchers are interested in modelling or finding the dynamic causal structure, such as for example between events in cellular reaction systems.
The following is a list of possible topics relevant to this workshop:
- Fundamentals of causal inference from longitudinal / time-dependent data
- Methods for dealing with time-varying confounders: inverse probability weighting, structural nested model, g-computation, Bayesian predictive inference
- Optimal dynamic decision strategies
- Control theory and machine learning
- Continuous-time dynamic modelling
- Applications, such as duration and timing of therapy for cancer or HIV patients, controlling of blood index by anticoagulant treatment, maintenance therapies for chronic conditions
The workshop is organised by Vanessa Didelez, Robin Henderson and Will Havercroft. You can contact them by email at
time-causality-organisers 'AT' sympa.bristol.ac.uk.
Confirmed Invited Speakers
| Prof Odd Aalen, Oslo |
Dynamic causal models |
| Prof Elja Arjas, Helsinki |
Why not simply apply stochastic process modeling and Bayesian predictive inference? |
| Dr Clive Bowsher, Bristol |
Stochastic Kinetic Processes: Biochemical Mechanisms,Information and Network Design |
| Prof Philip Dawid, Cambridge |
Applications of Dynamic Models in Monitoring and Fault Detection |
| Dr Roger Dixon, Loughborough |
A formal treatment of sequential ignorability |
| Dr Damien Ernst, Liège |
Learning for exploration-exploitation in reinforcement learning |
| Prof Erica Moodie, Montreal |
Q-learning for estimating optimal dynamic treatment rules from observational data |
| Prof Susan Murphy, Michigan |
Confidence Intervals, Q-Learning and Dynamic Treatment Regimes |
| Dr Susanne Rosthoj, Copenhagen |
Estimation of optimal dynamic treatments with irregular/missing visits using regret regressions |
| Prof Andrea Rotnitzky, Harvard |
Estimation and Extrapolation of Optimal Treatment Part 1 |
| Prof James Robins, Harvard |
Estimation and Extrapolation of Optimal Treatment Part 2 |
| Prof Jonathan Sterne, Bristol |
Causal inference for dynamic treatment regimens: how analyses of observational data changed international guidelines on when to start antiretroviral therapy |
| Dr James Taylor, Lancaster |
Control System Design: Feedback and Sequential Decision Making |
|
