Number of participants: 30
Instructors: Prof. Dr. Carsten Trenkler
Method (hours per week): lecture (2) + exercise (2)
Course level: Master
Prerequisites: E601 - E603 (or equivalent courses for exchange students)
Course Description: The lecture gives an introduction to multiple time series techniques and will cover vector autoregressive (VAR) processes, VAR estimation, VAR order selection and model checking. If time permits, we will also cover so-valled VEC models and Structural VAR models. The use of VAR models in forecasting, causality and impulse response analysis will be explained and illustrated using empirical examples and by discussing a selected set of research papers. The methods will be applied in computer tutorials.
Expected Competences acquired after completion of the module:
At the end of the term, students should be familiar with basic properties of time series processes and understand when it is beneficial to analyze multiple time series data. The main achievement of course participants will be to understand the methodological aspects of specifying and estimating multiple time series models and how they are used in forecasting. These tools are applied in empirical tutorials in the PC-Lab.
Contact person: Prof. Dr. Carsten Trenkler, E-Mail: mannsperger(at)uni-mannheim.de, L7, 3 - 5, Raum 107, Tel. 181-1845 (Sekretariat)