Course
Causal Inference in Empirical Economics - 5 ECTS
We, as human beings, tend to attribute cause and effect to observations quite quickly, even if such a causal relationship does not really exist. Causal inference is the science of the study of causal relationships and gives us tools to study rigorously if an intervention, action, or treatment actually causally determines a certain outcome. Causal inference is part of the answer to questions such as “What is the impact of heat on women’s reproductive health?”, “What is the effect of minimum wages on employment?”, or “To what extent do increases in food prices increase conflict?”
Lecturers
Dr. Elena Fumagalli, Dr. Karlijn Morsink (both Utrecht University) and Dr. Mark Treurniet (University of Groningen)
Introduction
In this “Causal Inference in Empirical Economics” course development economists from Wageningen University (WUR), and Utrecht University (UU), will teach state-of-the-art causal inference methods for both experimental and quasi-experimental designs, and help students to apply these to their own research designs.
The course will be taught fully online.
Learning objectives
After successful completion of this course, participants are expected to be able to:
Use economic theory to design a (quasi) experiment
- Apply and evaluate statistical techniques in terms of valid causal inference
- Appraise various experimental design choices
- Appraise various quasi-experimental methods
Course entry requirements
● DEC-32806 Impact Assessment of Policies and Programmes at WUR or ECRMRS1 Econometric Methods 1 at UU or a similar course at another university, and
● YSS-34306 Advanced Econometrics at WUR or ECRMRS1 Econometric methods 2 + Research skills: Data handling at UU or a similar course at another university, and
● Being able to program in Stata.
Activities
- Lectures: The course material will be discussed in seven interactive lectures
- Paper presentations: Students will present relevant papers on the state-of-the-art in causal inference in eight interactive lectures
- Assignments: Students will replicate some empirical results from highly-cited papers by implementing estimation procedures in Stata.
Feedback
● Question hours to discuss progress with Stata assignments.
● Each week, Q&A to help students apply the material and to ask for feedback on the Stata assignments.
Examination
The student needs to pass each:
A. Two Stata assignments (pass/fail grading)
B. Written take-home exam with open questions (minimum 5.5 to pass)
Tentative schedule
| Week 1: From Theory to design | ||
| 2-9 | 9.30 | Class 1 |
| 3-9 | Class 2 | |
| Week 2: One-sided and two-sided non-compliance | ||
| 9-9 | 9.30 | Class 3 |
| 10-9 | 9.30 | Class 4 |
| Week 3: Experimental research design | ||
| 15-9 | 9.30 | Class 5 |
| 17-9 | 9.30 | Class 6 |
| Week 4: Experimental research design | ||
| 22-9 | 9.30 | Class 7 |
| 24-9 | 9.30 | Class 8 |
| Week 5: Quasi-experimental methods I | ||
| 29-9 | 9.30 | Class 9 |
| 1-10 | 9.30 | Class 10 |
| Week 6: Quasi-experimental methods II | ||
| 6-10 | 9.30 | Class 11 |
| 8-10 | 9.30 | Class 12 |
| Week 7: Quasi-experimental methods III | ||
| 13-10 | 9.30 | Class 13 |
| 15-10 | 9.30 | Q&A |
| Week 8: Exam | ||
| 20-10 | Take-home exam |
More details on the programme can be found in the course manual.
Course fee
| WGS PhDs with TSP, UU PhDs, UG PhDs, WU MSc, UU MSc | € 0 |
| a) All other PhD candidates b) Postdocs and staff of the above mentioned Graduate Schools | € 650 |
| All others | € 975 |