Course in event history analysis
Language of instruction
The course will give an introduction to event history analysis and focus on extended event history analyses, such as competing risk analysis and multi-state models.
- Hazard function and survival function
- The Kaplan-Meier estimator
- Time dependent covariates
- The principle of competing risk
- Fine&Gray regression
- Multi-state data
- Data structures in extended event history analyses
- Pseudo data in event history analyses
At the end of the course, the student should have the following learning outcomes:
The student recognises complex event history data and set up data structures and models for extended event history analyses.
- arranges data files and performs some analyses in the statistical packages "R" or "Stata".
- understands some pros and cons of different models.
- understands the general principles of extended event history analyses.
- has a basis for work on extended event history data.
Every 3rd Spring
2 ECTS Credits. Comprises of 4 full day's seminars, 6 hours each.
Knowledge in medical statistics equivalent to MEDSTA, HELSTA or MEDSTAT1A is required.
Recommended previous knowledge
Seminars and exercises under guidance. Requirement of 80% attendance at the seminars to be entitled to an assessment.
Form of assessment
Home exam. All support material allowed. Evaluated as "Passed/ Not passed".
Who may participate
Study counsellor can be contacted here:
Phone: 55 58 60 16
The studyprogram comitee, Faculty of Medicine.
Course responsible: Prof. Stein Atle Lie, Stein.Lie@uib.no .
- David Collett: "Modelling survival Data in Medical Research, 3rd ed."