Regression Models and Graphs

Contact: wermuth@chalmers.se Course number: None; PhD course but interested others are welcome. Class times: Wednesdays 12:45 to 14:30 Fridays 13:00 to 14:45 last lecture is planned for 29. Oct 2010 Class rooms: Friday 8, 15, 29 in MVH12, 22 in MVH11 Wednesday 13, 20 in MVH11, 27 in MVH12 Theme of the course: Sequences of regressions in  joint or single responses provide a framework for assessing pathways of dependence, as they develop over time. Such pathways may often be modelled locally and  it may become simpler to understand them when a graph captures corresponding  conditional independences. It is the goal to see when and why this happens and how the recent results on regressions with independence structures can be exploited in analysing developmental data and in planning follow-up studies. Topics of the course Motivating examples from observational and intervention studies Regression models for discrete, continuous and mixed responses Checking for nonlinear and interactive effects Definitions and properties of independence graphs Markov equivalence and its consequences Constructing graphs from data and modifying them Causes for distorted dependences and avoiding them in follow-up studies Integrating results from related studies Password protected section (if it does not work please contact webmaster) Includes: reading, data and exercises.