Time Series Analysis

online course

Learn how to understand and apply various time series analysis methods for answering research questions in quantitative social science.

Need to know

You must have basic knowledge in statistical analysis including linear regression models and hypothesis testing and have a basic understanding of matrix algebra.

If you do not have this knowledge, take Introduction to Inferential Statistics or Applied Regression Analysis.
Prior familiarity with Stata or R is also required. Practical examples will be performed in either Stata or R.

In depth

Day 1

• What is a time series? Relationship to cross-sectional and panel data
• Stochastic processes, stationarity, autocorrelation functions
• ARMA processes

Day 2

• Autoregressive Distributed Lag (ADL) models
• Error Correction (EC) models
• Impulse responses and variance decompositions
• Vector Autoregressive (VAR) Processes
• Granger Causality

Day 3

• Unit roots and integrated process
• Spurious regression
• Event studies, structural breaks and regime-switching

Day 4

• Expanding the cross-section
• Panel Data and Multilevel models

Day 5

• Nonlinearity
• Aggregation issues
• State-space approach and Bayesian modelling

How the course will work online

The course combines pre-recorded lectures with daily two-hour live Zoom sessions. The pre-recorded lectures will provide an overview of relevant concepts and statistical foundations necessary to apply time series analysis. The Zoom sessions focus on two tasks:

• Discussion of questions students have.
• A lab session with hands-on exercises.

You will get to know each other and each other's projects in the first live session and explore how you can apply time series analysis to answer your research questions in social science. During each Zoom session, there will be exercises for you to complete, which will be discussed together. If you have any questions or thoughts to share, you can post them on Canvas.

Course leader

Chendi Wang is assistant professor in political science at VU Amsterdam. His research interests include political behaviour, political economy, comparative politics, and quantitative and computational methods.

Target group

Researchers, professional analysts, advanced students

Course aim

Cross-sectional statistical methods rely on the assumption that observations are drawn independently from a population. Observations that are observed over time are typically highly dependent, which leads to the breakdown of many traditional tools. The systematic approach towards analysing ordered, dependent data is called time series analysis. Examples of time series data are mostly macro (-economic, -sociological, -political) as well as many data in the environmental, physical, or medical sciences.

The aim of the course is to teach you how to understand and apply various time series analysis methods for answering research questions in quantitative social science. By the end of the course, you will:

• Understand the basics of time series data.
• Estimate and interpret the empirical autocorrelation function
• Estimate and compare models for stationary series
• Test for stationarity of time series data
• Compare and assess dynamic models
• Apply time series analysis to social science research questions

Credits info

4 EC
You can earn up to four credits for attending this course.
3 ECTS credits – Attend 100% of live sessions and engage fully with class activities.
4 ECTS credits – Attend 100% of live sessions, engage fully with class activities and complete a post-class assignment.

Fee info

GBP 478: ECPR Member
GBP 956: ECPR Non-Member

Scholarships

Funding applications for the 2023 ECPR Summer School in Research Methods and Techniques are now closed.
For more details on funding opportunities for ECPR's other events, please visit our website.