• Identify and describe key features of time series data such as trend, seasonality, structural breaks and outliers.
• Evaluate data quality and reliability, and reflect on practical challenges related to the data quality.
• Apply various time series forecasting models
• Assess model assumptions such as stationarity
• Compare forecasting models using relevant evaluation metrics. Justify the choice of model for a given forecasting problem and data.
• Evaluate a forecasting analysis conducted by another person/researcher.
• Interpret model output and forecasts.
• Communicate analysis and results clearly in both academic and policy context.
• Reflect critically on modelling decisions.
• Use R to implement forecasting models, document analysis, and present results in a transparent and reproducible manner.

This course provides a solid theoretical foundation for predictive modelling while placing strong emphasis on hands-on, applied analysis of economic and business data. Students work actively with real and synthetic time series in R, learning both how to produce forecasts and how to interpret, evaluate, and communicate the results in a clear and meaningful way.
 

Throughout the course, students should develop intuition for how to approach data: exploring and describing a time series, assessing data reliability and limitations, and identifying features such as trend, seasonality, outliers, and structural breaks. Understanding the data is a central step of the forecasting process, and the course integrates practical examples to highlight the types of challenges that arise in real-world settings.
 

The course covers several classes of forecasting models from different branches of statistics. In each case, the focus is on developing forecasting intuition: understanding why a model behaves as it does, the assumptions involved, when it is appropriate to apply a given method, and how to diagnose potential problems. Students learn to compare alternative models, justify their modelling choices, and critically evaluate forecasting performance using a range of diagnostic tools.

Predictive analytics plays a key role in managerial and policy decision-making.

The course therefore highlights practical applications, including forecasting sales, demand, and financial variables. Students learn how to prepare relevant data for forecasting, select well-suited models, interpret output, and communicate uncertainty to stakeholders. Tools for model selection with larger datasets are also introduced.

• Classic and basic theory
• Methodology
• Models

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

The literature can be changed before the semester starts. Students are advised to find the final literature on Canvas before they buy the books.

Main text book(s) and articles:

Hyndman, R.J & Athana­sopou­los, G (2021): Forecasting: principles and Practice   (HA) (Free online version available)

Gujarati, D.N. and Porter, D.C. (2010). Essentials of Econometrics, 4th edition, McGraw-Hill, pp 201-205 and 386-397

Doornik, J.A. & Hendry,  D.F.  (2014): Statistical model selection with ‘Big Data’. University of Oxford, Department of Economics, Discussion Paper Series,  # 735  (or more recent text in this spirit). (DH)

More supplementary stuff:

Buus Lassen, N., la Cour, L., Vatrapu, R. (2017), ‘Predictive Analytics with Social Media data’ in Sloan & Quan-Haase ed. The SAGE Handbook of Social Media Research Methods, Chapter 20, pp 328-341 (BCV)