Causal research and Prediction modeling

The intent of this intensive short course is to strengthen the methodological skills of the research community. Upon successful completion of the course, participants will have a deeper understanding of methods in causal and prediction research and increased confidence in how to apply these tools in their everyday research practice.

Lecturers:

• Rolf H.H. Groenwold, MD, PhD, Leiden University Medical Center (NL)
  Maarten van Smeden, PhD, University of Utrecht, Utrecht (NL)

Learning objectives:

By the end of this week, participants should be able to:

• Critically assess the results of epidemiological studies on causal relationships or prediction models

• Correctly define exposures and learn how to best represent them in models

• Understand difference between various sources of bias (confounding, measurement error and missing data) and the way these biases may differentially affect studies on causal relationships and prediction models.

• Describe key assumptions of methods used to control for (time-varying) confounding.

• Describe key assumptions of methods used to handle missing observations.

• Understand the reasons for and consequences of overfitting prediction models

• Describe recent developments in the fields of causal research and prediction modelling

Prerequisites:

• Basic knowledge of epidemiology

• Familiarity with R statistical software (for a short introduction see http://www.r-tutorial.nl/)

Fees:

• 750 €

• 510 € for enrolled students (proof required)
   

3 ECTS

Registration Information:
Tanja Te Gude
Tel. +49 30 450 570 812