The course provides an introduction to statistical theory and methodology for causal effect studies with different types of data. Application methods for solving problems in variety of fields, such as health, economics, public policy and education, are discussed.
The course is intended to Master’s degree or Ph.D. in statistics students and Ph.D. students from other disciplines who are interested in exploring the reasoning behind analytical approaches to answer causal questions.
Studies in statistical inference and regression analysis.
The course will alternate between lectures and discussions based on selected chapters from reference books and selected papers (2 + 2h/week).
Attendance: min 12/14 classes is required.
The course starts with causal thinking and its impact on scientific inquiry, particularly when answering causal questions. What is the definition of ‘causal’ in statistical terms and how it is justified will provide us with the thinking framework for introducing Potential Outcomes Approaches to Causal Inference with experimental and observational data. Design-based versus model-based methods will be discussed. Due to assumptions playing a major role in causal inference studies: “Causal inference without assumptions is impossible”, we will make an effort to understand this profound limitation and learn how to address it, also by looking at different approaches to Sensitivity Analysis.
- Freedman, D. (2009). Statistical Models and Causal Inference: A Dialogue with the Social Sciences. Cambridge: Cambridge University Press.
- Imbens, G., & Rubin, D. (2015). Causal Inference for Statistics, Social, and Biomedical Sciences: An Introduction. Cambridge: Cambridge University Press.
- Rosenbaum P. (2017). Observation and experiment. An introduction to causal inference. Cambridge: Harvard University Press.
Announced in the material section
Weekly assignments: (50%)
Final written assignment (50%)
min 12/14 classes is required.
Ana Kolar, PhD: https://www.tarastats.com/about/#anakolar