Navigated to
Syllabus:

Causal inference in register studies, 7.5 credits

The course is discontinued

Swedish name: Kausal inferens vid registerstudier
This syllabus is valid: 2014-12-29 valid to 2017-08-20 (newer version of the syllabus exists)
Course code: 2ST022
Credit points: 7.5
Education level: Second cycle
Main Field of Study and progress level: Statistics: Second cycle, has only first-cycle course/s as entry requirements
Grading scale: Pass with distinction, Pass, Fail
Responsible department: Department of Statistics
Revised by: Rector of Umeå School of Business and Economics, 2014-12-18

Contents

The course is an introduction to Causal Inference in Observational Studies.
Two theoretical frameworks are presented

  • Potential outcomes (Rubin Causal Model) and  
  • Structural models

The course covers how causal parameters can be identified in the two theoretical frameworks. The models and estimators presented are motivated by the need to control for confounding in observational studies. Non- and semi-parametric estimators of causal effects and their properties are described. Challenges and opportunities when register data is used in the analyses will also be discussed during the course.

 

Expected learning outcomes

After the course the student should be able to

  • describe the foundations of the Neyman-Rubin framework of causal inference
  • describe the basis of inference when the sample consists of the whole population
  • apply and explain directed acyclic graphical models
  • apply non- and  semi-parametric estimators of causal effects
  • estimate causal effects with statistical software
  • perform register-based evaluations (in combination with other sources of micro data.).

Required Knowledge

At least 75 credits in statistics and/or mathematical statistics, or equivalent. Proficiency in English equivalent to Swedish upper secondary course English B (English/6)
 

Form of instruction

The course consists of lectures and lessons. There are mandatory assignments where the students shall present their solutions.
 

Examination modes

The examination partly consists of individual written reports and oral presentations of given assignments. Written reports of assignments should be handed in or presented orally at predetermined dates. The grades for the assignments are: G (Pass), and U (Fail).
 
The examination also includes a comprehensive individual oral exam. The grades for the oral exam are: G (Pass), U (Fail), VG (Pass with distinction).
 
Grades on the course are awarded when students have passed all examinations in the course. The gr­ade is a com­prehensive eva­luation of the results of the various parts of the examinations and is not granted until all mandatory tasks have been passed.
 
A student who has passed an examination is not allowed to take another examination in order to get a higher grade. For students who do not pass, an additional test will be held according to a pre-determined schedule.

After two failed examinations on the course, the student has the right to request another grading teacher. Writ­ten requests should be handed to the Director of Studies no later than two weeks before the date of the next exam­ination.

Examinations based on the same course syllabus as the ordinary examinations are guaran­teed to be offered up to two years after the date of the student's first registration for the course.

Academic credit transfer

Academic credit transfers are according to the University credit transfer regulations.
 

Literature

Valid from: 2015 week 1

Recent developments in the econometrics of program evaluation
Imbens G.W., Wooldridge J.M.
Included in:
Journal of economic literature.
Nashville, Tenn. The Assoc. : 1969- : 47 : 5-86 pages
Mandatory

Vansteelandt S
Estimating direct effects in cohort and case-control studies
Included in:
Epidemiology.
Cambridge, MA : Blackwell Scientific Publications : 1990- : 20 : 851-860 pages
Mandatory

An overview of relations among causal modeling methods
Greenland S., Brumback B.
Included in:
International journal of epidemiology
Oxford : Oxford U. P. : 1972- : 31 : 1030-1037 pages
Mandatory

Marginal structural models for analyzing caulas effects of time-dependent treatments: an application in perinatal epidemiology.
Bodnar L.M., Davidian M. et al
Included in:
American journal of epidemiology
Baltimore, Md. : 1965- : 926-934 pages
Mandatory

Stratification and weighting via the propensity score in estimation of causal treatment effects: a comparative study
Lunceford J.K., Davidian M
Included in:
Statistics in medicine.
Chichester : Wiley : 1982- : 2937-2960 pages
Mandatory

Pearl Judea
Causality : models, reasoning, and inference
2. ed. : Cambridge : Cambridge University Press : 2009 : xix, 464 p. :
ISBN: 052189560X (inb.)
Mandatory
Search the University Library catalogue

Rosenbaum Paul R.
Observational studies
2nd ed. : New York : Springer : c2002. : xiv, 375 p. :

ISBN: 0-387-98967-6 (acid-free paper)
Mandatory
Search the University Library catalogue