Магистратура 2020/2021

Байесовская статистика

Лучший по критерию «Полезность курса для расширения кругозора и разностороннего развития»

Лучший по критерию «Новизна полученных знаний»

Статус: Курс обязательный (Прикладная политология)

Направление: 41.04.04. Политология

Кто читает: Департамент политики и управления

Где читается: Факультет социальных наук

Когда читается: 2-й курс, 1, 2 модуль

Формат изучения: без онлайн-курса

Преподаватели: Маркварт Кайл Лоус

Прогр. обучения: Прикладная политология

Язык: английский

Кредиты: 6

Контактные часы: 56

Course Syllabus

Abstract

In this course, students will learn both basic theory of Bayesian statistical analysis as well as important applications that use Bayesian methods. Specifically, the course will focus on using Bayesian methods to analyze survey data and latent concepts, as well as data with a hierarchical structure. Prerequisites: “Methodology and research methods of political science” and “Methods of analyzing heterogeneous data”.

Learning Objectives

• Understanding and applying Bayesian methods with important applications; using these methods to produce innovative research outputs.

Expected Learning Outcomes

• Students will be able to design and interpret Bayesian statistical analyses, using prominent software such as JAGS and STAN.
• Students will learn how to design and implement Bayesian multilevel models and measurement models.

Course Contents

• COURSE PLAN
  The course will consist of 14 classes, each consisting of a lecture and seminar. Most classes will be divided into 1) a lecture in which students will be introduced to the theoretical and practical issues involved in a method, and 2) a seminar in which students will work with the instructor and small groups to implement the techniques to which they were exposed. There will also be two days of in-class presentations and one day for collaborative work on projects. The class will be divided into four thematic groups, each with two or more lectures: 1) an introduction to Bayesian statistics, 2) Bayesian regression analysis, 3) hierarchical models, and 4) latent variable models.
• Class 1-2: Introduction to Bayesian statistics
• Class 3-6: Bayesian regression analysis
• Class 7: Model choice / Simulation analyses
• Class 8: In-class presentations of replication analyses
• Class 9-10: Hierarchical models
• Class 11-13: Latent variable models
• Class 14: Final presentations

Assessment Elements

• Participation
• Homework
• Replication project
• Final paper

Interim Assessment

• Interim assessment (2 module)
  0.35 * Final paper + 0.1 * Homework + 0.2 * Participation + 0.35 * Replication project

Bibliography

Recommended Core Bibliography

• Daniel Pemstein, Stephen Meserve, & James Melton. (2010). Democratic Compromise: A Latent Variable Analysis of Ten Measures of Regime Type.” Political Analysis 18(4):426.
• Simon Jackman. (2009). Bayesian Analysis for the Social Sciences. Wiley.

Recommended Additional Bibliography

• Gelman, A., Carlin, J. B., Stern, H. S., Dunson, D. B., Vehtari, A., & Rubin, D. B. (2014). Bayesian Data Analysis (Vol. Third edition). Boca Raton: Chapman and Hall/CRC. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=nlebk&AN=1763244
• Shawn Treier, & Simon Jackman. (2008). Democracy as a latent variable.