Master
2024/2025
Statistical Analysis II
Category 'Best Course for Career Development'
Category 'Best Course for Broadening Horizons and Diversity of Knowledge and Skills'
Category 'Best Course for New Knowledge and Skills'
Type:
Elective course (Population and Development)
Area of studies:
Public Administration
Delivered by:
Department of Higher Mathematics
Where:
Faculty of Social Sciences
When:
1 year, 3, 4 module
Mode of studies:
offline
Open to:
students of one campus
Instructors:
Daria Salnikova
Master’s programme:
Population and Development
Language:
English
ECTS credits:
6
Course Syllabus
Abstract
This course offers a practical introduction to modern applied statistics and econometrics, assuming that participants already have a basic knowledge of statistical inference, and linear regression models. The teaching approach follows a clear two‑step logic: each mathematical concept is first introduced intuitively and then formalized, after which it is immediately put into practice on real‑world problems.
The curriculum is built around three core modules: 1) regression models with interaction terms
2) logistic regression models for binary outcomes
and 3) an introduction to panel data analysis (covering basic specifications and estimation techniques)
Throughout the course, Python is used as the primary computational tool, enabling students to apply statistical methods to authentic datasets, visualize results, and interpret outputs.
Learning Objectives
- The goal of this course is to improve students’ skills in the linear regression analysis, to learn how to estimate the model with the binary dependent variable, to learn how to estimate FE and RE panel models, learn how to estimate difference-in-differences model, to make students familiar with the basic tools for testing theories, to make students able to read, interpret and replicate the results of published papers using Python and real-world data
Expected Learning Outcomes
- Be able to use theoretical notions, concepts and interpret the models with Panel Data.
- to learn how to estimate the model with the binary variable
- Explain the difference between fixed-effect, random-effects, and first-difference models; the parallel trends assumption
- Be able to address endogeneity problems
- Know properties of maximum likelihood estimates.
- be able to identify cases when it is possible to use IV regression models
- be able to estimate the IV regression model
- be able to define and use the maximum likelihood estimation approach
- be able to apply difference-in-differences model
- be able to interpret the difference-in-differences model
- be able to identify a strong and a weak instrument
- to be able to analyze and estimate Panel Data models on real data
- Able to run regression models with interaction terms
Course Contents
- Regression models with interaction terms
- Panel Data Models
- Difference-in-Differences
- Endogeneity. Instrumental variables method. 2SLS
- Maximum Likelihood Estimation
- Binary dependent variables. Logit and probit models
Assessment Elements
- Test 1
- Home assignment 1
- Test 2:-
- Home assignment 2
- Exam
- Seminar Activity
- Home Assignment 3
Interim Assessment
- 2024/2025 4th module0.05 * Home Assignment 3 + 0.05 * Home assignment 2 + 0.45 * Exam + 0.15 * Test 1 + 0.1 * Seminar Activity + 0.15 * Test 2: + 0.05 * Home assignment 1
Bibliography
Recommended Core Bibliography
- Counterfactuals and causal inference : methods and principles for social research, Morgan, S. L., 2015
- Data analysis using regression and multilevel/hierarchical models, Gelman, A., 2009
- Introduction to econometrics, Stock, J. H., 2008
Recommended Additional Bibliography
- Beck, V. L. (2017). Linear Regression : Models, Analysis, and Applications. Hauppauge, New York: Nova Science Publishers, Inc. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsebk&AN=1562876