Causal Inference and Policy Evaluation**
A.Y. 2024/2025
Learning objectives
The main objective of this course is to introduce students to the concepts of causality and counterfactual impact evaluation (CIE). After a statistical introduction to the most popular methods used to assess causality (such as instrumental variables, difference-in-differences, regression discontinuity design, randomized control trials), their application will be illustrated through examples of causal inference and policy evaluation in several domains of applied economics, such as labor, education and health economics.
Expected learning outcomes
After the end of the course students must be able: 1) to read and understand the specialized literature using causal inference; 2) facing a problem of causal inference, to understand the complexities involved and to carefully design research strategies to estimate causality and evaluate policies; 3) to carry out policy evaluations in most fields of applied economics, to write a report (or paper) explaining the results, and to present them to both an academic and a general audience.
Lesson period: First trimester
Assessment methods: Esame
Assessment result: voto verbalizzato in trentesimi
Single course
This course can be attended as a single course.
Course syllabus and organization
Single session
Responsible
Lesson period
First trimester
Course syllabus
The counterfactual and causality
Recap of Ordinary Least Squares (OLS) and causality
Randomized Control Trials (RCTs)
Instrumental Variables (IVs)
The Local Average Treatment Effects (LATE) of IVs
Difference-in-Differences (DID)
Regression Discontinuity Design (RDD)
Recent developments of causal inference and policy evaluation methods
Recap of Ordinary Least Squares (OLS) and causality
Randomized Control Trials (RCTs)
Instrumental Variables (IVs)
The Local Average Treatment Effects (LATE) of IVs
Difference-in-Differences (DID)
Regression Discontinuity Design (RDD)
Recent developments of causal inference and policy evaluation methods
Prerequisites for admission
Students are expected to be familiar with the basic concepts taught in undergraduate-level statics and/or econometrics courses, including multivariate regression analysis, and be able to use statistics/econometrics software (e.g. R, STATA). The software used in the course is STATA.
Students are expected to be already familiar with demand and production theory, acquired in the "Advanced Microeconomics and Macroeconomics" (DSE, MEF) or "Advanced Microeconomics" (EPS) courses, and with the main empirical strategies used to assess causality in econometrics (i.e. taught in the DSE course "Micro-econometrics, Causal Inference and Time Series Econometrics" or the EPS course "Empirical Methods for Economics and Policy Evaluation"). Knowledge of statistical/econometric software packages such as R or STATA is needed to complete the (voluntary but strongly recommended for attending students) paper assignment and class assignments. (Some basic knowledge of the STATA software can be acquired through the EPS "Advanced Computer Skills" course, from the 2020-21 edition.)
Students are expected to be already familiar with demand and production theory, acquired in the "Advanced Microeconomics and Macroeconomics" (DSE, MEF) or "Advanced Microeconomics" (EPS) courses, and with the main empirical strategies used to assess causality in econometrics (i.e. taught in the DSE course "Micro-econometrics, Causal Inference and Time Series Econometrics" or the EPS course "Empirical Methods for Economics and Policy Evaluation"). Knowledge of statistical/econometric software packages such as R or STATA is needed to complete the (voluntary but strongly recommended for attending students) paper assignment and class assignments. (Some basic knowledge of the STATA software can be acquired through the EPS "Advanced Computer Skills" course, from the 2020-21 edition.)
Teaching methods
Frontal lectures.
Teaching Resources
- Angrist, J. D., & Pischke, J. S. (2008). Mostly harmless econometrics. In Mostly Harmless Econometrics. Princeton university press.
- Scientific papers posted on the ARIEL website.
Non-technical introduction to the course topics
Angrist, J. D., & Pischke, J. S. (2015). Mastering 'Metrics. The path from cause to effect. Princeton University Press.
- Scientific papers posted on the ARIEL website.
Non-technical introduction to the course topics
Angrist, J. D., & Pischke, J. S. (2015). Mastering 'Metrics. The path from cause to effect. Princeton University Press.
Assessment methods and Criteria
Attending students
Attending students have to write an empirical paper (either single-authored or in groups of 2-3 students) using causal inference and policy evaluation methods (e.g. evaluating a given policy) in which they use the econometric methods learned during the course.
=================================
Non-attending students
The course is assessed through a final written exam, which mainly consists of open and multiple-choice questions. The main emphasis of the course is on developing students' abilities to formulate a research question aiming at assessing causality and writing a research paper. Thus attendance is strongly advised.
Attending students have to write an empirical paper (either single-authored or in groups of 2-3 students) using causal inference and policy evaluation methods (e.g. evaluating a given policy) in which they use the econometric methods learned during the course.
=================================
Non-attending students
The course is assessed through a final written exam, which mainly consists of open and multiple-choice questions. The main emphasis of the course is on developing students' abilities to formulate a research question aiming at assessing causality and writing a research paper. Thus attendance is strongly advised.
Educational website(s)
Professor(s)
Reception:
Unless otherwise notified (published here): Tuesday 18:00-19:30; Wednesday 18:00-19:30. On appointment.
MS Teams or in person (office nr. 21 DEMM)