To illustrate possible topics for Bachelor and Master theses at our chair, this page lists a selection of completed theses. For all projects, we require relevant statistical knowledge (as taught in our courses, for example) as well as very good programming skills in R or Python. The choice of topics is based on the previous knowledge and interests of the students, as well as on our current research topics at the chair. If you are interested, please contact the administration office with your current transcript of records and a short CV.
- Generating Predictive Distributions for Energy Market Data
- Multivariate Forecast Distributions for European Economic Indicators
- Monitoring Forecast Performance in the Presence of Concept Drift
- Predicting Return Distributions via Random Forests
- Prune to Inform: Feature Importance with Supermasks
- Quantifying Forecast Uncertainty by Turning Point Forecasts into Probability Distributions
- Scoring Rules, Energy Distance und Kernel-basierte Zweistichproben-Tests: Eine vereinheitlichende Betrachtung
The formal requirements for a seminar paper or thesis are summarized here.