Магистратура
2026/2027
Эконометрические методы причинно-следственного анализа
ID 1164767
Статус:
Курс обязательный (Аналитика данных для бизнеса и экономики)
Кто читает:
Департамент экономики
Где читается:
Санкт-Петербургская школа экономики и менеджмента
Когда читается:
1-й курс, 2, 3 модуль
Охват аудитории:
для своего кампуса
Преподаватели:
Шагбазян Гегам Варданович
Язык:
английский
Кредиты:
6
Контактные часы:
60
Course Syllabus
Abstract
The course includes the study of the following range of issues: binary choice models, multiple choice models, models with countable variables, Tobit models, duration models, impact models. The main features of the course are the illustration of theoretical material with examples of empirical assessments using computer programs and the acquisition of computer skills in econometric packages by the course students.
Learning Objectives
- The course aims to: equip students with a rigorous understanding of core econometric methods for causal inference with limited dependent variables and treatment effect models; develop the ability to distinguish between correlation and causation and to select appropriate identification strategies for empirical research questions; provide practical skills in implementing these methods using modern econometric software; prepare students for independent empirical research in business analytics and applied economics.
Expected Learning Outcomes
- formulate a causal research question using the potential-outcomes framework and clearly define the treatment, outcome, counterfactual, and causal parameter of interest;
- explain why the comparison of treated and untreated units may not identify a causal effect and identify the sources of selection bias and confounding;
- derive and interpret the Average Treatment Effect and Average Treatment Effect on the Treated and explain the assumptions under which these parameters can be estimated;
- evaluate a randomised experiment by explaining the role of random assignment, treatment compliance, attrition, and differences between intention-to-treat and treatment-on-the-treated effects;
- explain the logic of matching and propensity-score methods and assess whether the assumptions required for matching are plausible in a given empirical application;
- сritically read an empirical economics paper by identifying its causal estimand, source of identifying variation, assumptions, estimation strategy, and main threats to identification;
Course Contents
- 1. Causal Inference and the Counterfactual Framework
- 2. Randomised Experiments
- 3. Selection Bias and Confounding
- 4. Regression Adjustment and Control Variables
- 5. Matching and Selection on Observables
- 6. Treatment-Effect Heterogeneity
- 7. Instrumental Variables
- 8. Local Average Treatment Effects and Weak Instruments
- 9. Regression Discontinuity Designs
- 10. Difference-in-Differences
- 11. Panel Data and Fixed Effects
- 12. Inference and Robustness in Causal Research
Interim Assessment
- 2026/2027 3rd module0.25 * Midterm test + 0.5 * Final presentataion + 0.25 * Midterm test
