GOALS AND RATIONALE
Causal inferences play a predominant role in science.
In epidemiology, the goal and the ambition of the most part of the researchers is to determine an unbiased estimate of the effect of being exposed to a given risk factor on a well defined outcome (disease, death). In recent years, there have been important statistical developments that go be-yond the traditional multivariable regression techniques in order to obtain unbiased estimates.
Aims of this course are to discuss the current state of the art with respect to these issues, while retaining a practical focus and to assess our current and future abilities to ad-dress effectively cause-and-effect questions.
10 December 2014 – 9:00/18:00
Basic concepts in epidemiology seen through causal infer-ence and causal diagrams: measures of association, bias, confounding, missing values. Instrumental variables.
11 December 2014 – 9:00/18:00
Mediation analysis. Interaction and effect Modification
12 December 2014– 9:00/17:30
Causal estimation methods: Introduction to marginal structural models for fixed and time-varying confounders
Teaching will be based on both formal lectures and com-puter/group sessions.
Seminar Room n. 2062, Department of Statistics and Quantitative Methods, University of Milano-Bicocca, Via Bicocca degli Arcimboldi 8, Building U7 (second floor), 20126 Milan (Italy).
For further information see the brochure. For application, please visit the website http://www.causal.altervista.org/courses.php.
For information on “how to reach us” please visit the page http://biostatunimib.altervista.org/reach-us/ .