Магистратура
2026/2027
Дизайн и анализ экспериментов
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, 3 модуль
Охват аудитории:
для всех кампусов НИУ ВШЭ
Преподаватели:
Афанасьев Кирилл Олегович
Язык:
английский
Кредиты:
3
Контактные часы:
32
Course Syllabus
Abstract
Business, academic research, and public administration often rely on experiments. New ideas, products, or policies are typically tested on a subset of the target population before being implemented at scale. In academic research, experiments are used to identify causal relationships.
Unlike the econometric approach, which focuses on answering research questions using existing data, the experimental approach focuses on designing the environment in which data will be collected and determining how those data should be collected to answer a research question. Accordingly, this course addresses three fundamental questions: How should an experiment be optimally designed? How should experimental data be properly analyzed? And how should the results be adequately interpreted?
Learning Objectives
- Formation of an understanding of the foundations of experimental research and causal inference.
- Development of analytical thinking through the interpretation and evaluation of experimental evidence.
Expected Learning Outcomes
- Use potential outcomes approach or other frameworks to analyze causal effects of different interventions.
- Calculate optimal sample sizes
- Select appropriate statistical methods for experimental data
- Select appropriate experimental types
- Use appropriate experimental methods to analyze moderation and mediation
- Design optimal experiments with longitudinal elements
- Design optimal within-subject experiments
- Analyze spillovers in experiments
- Interpret results of economic experiments
- Discuss horizontal and vertical generalizabillity of economic experiments
- Analyze missing data
- Maximize compliance in experiments
- Handle data with compromised randomisation
Course Contents
- Introduction to experiments in economics
- Basics of experimental design
- Moderation and mediation
- Experiments with longitudinal elements
- Within-subject experiments
- Issues in experimental data
- Confidence, generalizability, scaling
Assessment Elements
- In-class activity
- Online tests
- Experimental pre-registration presentation
- Midterm exam
- Final exam
Interim Assessment
- 2026/2027 3rd module0.25 * Experimental pre-registration presentation + 0.3 * Final exam + 0.1 * In-class activity + 0.1 * Online tests + 0.25 * Midterm exam
Bibliography
Recommended Core Bibliography
- Natural experiments in the social sciences : a design-based approach, Dunning, T., 2012
Recommended Additional Bibliography
- Causal inference in statistics, social, and biomedical sciences : an introduction, Imbens, G. W., 2020
- Introduction to mediation, moderation, and conditional process analysis : a regression-based appr..., Hayes, Andrew, F., 2018