• Formulate a research question that draws on real-world time series data from economics, business, or finance.
• Answer your research question using time series models.
• Identify, select, and defend the choice of data and analytical method that best fit your research question.
• Explain your question’s contribution to the field, its societal importance, and the implications of your analysis for decision-making in business, economics, or finance.
• Use R to conduct a full time series analysis, from importing and cleaning data to estimation, visualization, and presenting results.
• Understand the theoretical foundations and assumptions behind key time series models, and critically evaluate the strengths and limitations of different approaches.
• Interpret and communicate results from time series models, including parameter estimates, significance tests, and forecast intervals

What we cover in this course will be excellent practice for writing your master thesis and will prepare you to be well‑equipped for roles in analytics departments across industries.

You will become proficient in R—a prevalent and often required software in financial institutions, firms, and academia—and you will learn to use it effectively for both study and work. Step by step, you will learn how to analyze time series data, work with univariate and multivariate datasets, build and evaluate forecasts, establish long‑run relationships (e.g., via cointegration/ECM), and analyze responses to shocks (e.g., impulse‑response analysis and volatility dynamics).

In the first two weeks, we’ll work with self‑simulated data to build intuition for the fundamentals of time series analysis—stationarity, unit roots, and random walks—without the distractions of messy real‑world data.

From the third week onward, we’ll switch to real‑world data, downloaded in real time. Each week, I’ll briefly introduce the theory behind a model, and then we’ll spend most of the class applying that model to actual data. If you want to go deeper, I’ll share optional materials with full proofs and derivations. Lecture notes will be available in PDF, HTML, and Jupyter Notebook formats.

Every class will begin with a short Slido exercise to check understanding and to make it easier and more natural for you to ask questions. After that, you’ll have one or two in‑class exercises (about 20 minutes each), which you can do individually, in pairs, or in small groups—whichever you prefer.

From day one, I’ll encourage you to form pairs or  groups of 3 or 4 and start working on your course project. If you can’t find a group on your own, I will facilitate group formation so that everyone has a chance to collaborate. The project—first a written synopsis and then an oral examination—is designed to mimic the master thesis process and oral defense. It’s excellent practice for the critical thinking and diligence you’ll need for your thesis.

How the project fits into the weekly structure

• Weeks 1–2: Fundamentals (stationarity, unit roots, random walks). You’ll practice with simulated data and start drafting research questions for your project.
• Week 3: ARIMA models. You’ll estimate your first models and consider how they fit your project.
• Week 4: ADL (autoregressive distributed lag) models. You’ll expand your modeling toolbox.
• Week 5: Forecasting with ARIMA and ADL. You’ll learn how forecasting enters your project.
• Weeks 6-7: VAR (vector autoregression) models. You’ll model interactions across multiple variables.
• Week 8: ECM (error correction models) and cointegration. You’ll connect short‑run dynamics with long‑run relationships.
• Week 9: ARCH/GARCH models. You’ll learn to model and interpret volatility—especially useful for financial/business data.
• Week 10: Exam preparation.

Project Guidelines

Your project should look and feel like a mini‑research paper.

Follow this structure:

1. Introduction
   • State your research question.
   • Explain why it’s worth studying—why should anyone care?
   • Summarize the literature/practice and highlight what is new in your approach.
2. Data
   • Describe the source, frequency, seasonal adjustment, and time range; justify your choices.
   • Provide figures/visualizations and comment on missing data if relevant.
   • Test for stationarity: state null/alternative hypotheses, test statistics, significance levels, and critical values; complement with visual inspection and interpretation.
   • Document and justify any transformations you apply to obtain stationarity.
3. Methods
   • Explain which models (ARIMA, VAR, ECM, etc.) you use, how, and why.
   • Describe the diagnostic checks (e.g., serial autocorrelation tests, recursive estimation, out‑of‑sample forecasting) and their purpose.
4. Estimation
   • Present results: discuss fit, parameter significance, signs/magnitudes, and how they compare to expectations and the literature.
   • Explain your treatment of insignificant coefficients (keep, drop, or re‑estimate) and why.
   • Report diagnostics and any model adjustments; if adjustments aren’t made, explain implications.
5. Conclusion and Interpretation
   • Identify your best model(s) and use them to answer your question.
   • Relate findings to the literature and business/policy practice.
   • Reflect on the usefulness of your results for policymakers, business planners, or other decision‑makers.

At the end of the course, you’ll present your findings in a written report and an oral defense. Think of this as a trial run for your master thesis—your chance to practice being critical, thorough, and creative in your research.

• Classic and basic theory
• Methodology
• Models

• Development of research questions
• Data collection
• Analysis
• Discussion, critical reflection, modelling
• Peer review including Peer-to-peer
• Activities that contribute to new or existing research projects
• Students conduct independent research-like activities under supervision

Applied Econometric Time Series - Walter Enders, 4th Edition, Wiley