Methods of Longitudinal Data Analysis 2400-ZEWW1037

The course introduces two methods of longitudinal data analysis: sequence analysis and event history analysis (survival analysis). These methods are widely used in demography, epidemiology, and economics (e.g., labor economics and risk assessment) and allow researchers to address questions such as: how do women’s and men’s career trajectories evolve over the life course, what factors determine the risk of job loss or firm bankruptcy, and what influences the timing of first childbirth.
The course consists of lectures and computer laboratories during which students will learn how to prepare longitudinal data, conduct basic descriptive analyses, and estimate hazard models using Stata. Students will also receive materials introducing basic model estimation methods in R. At the end of the course, students will complete a mini-project based on the analysis of microdata.
The course covers the following topics:
1. What are longitudinal data and longitudinal data analysis? Examples of applications in demography and economics
2. Sequence analysis: general concepts, sequence visualization, measuring similarities and differences between sequences, sequence clustering, interpretation of results, practical applications, introduction to sequence analysis in Stata
3. Introduction to event history analysis
4. Event history models: exponential and piecewise constant models, parametric models, Cox proportional hazards model
5. Specification of hazard models: time-varying covariates, interactions, anticipatory analysis, model diagnostics
6. Event history analysis in discrete and continuous time
7. Practical exercises: data preparation, model estimation, diagnostics, presentation and interpretation of model results
Classes will be conducted in Stata, and students will also receive basic guidance on estimating basic models in R.