Informations about the modules

zurück

Module (6 Credits)

Energy Forecasting Competition

Name in diploma supplement
Energy Forecasting Competition
Responsible
Admission criteria
See exam regulations.
Workload
180 hours of student workload, in detail:
  • Attendance: 60 hours
  • Preparation, follow up: 100 hours
  • Exam preparation: 20 hours
Duration
The module takes 1 semester(s).
Qualification Targets

The students

  • learn concepts to produce and evaluate probabilistic forecasts
  • can produce forecasts using python or R for time series data from energy systems and markets
  • learn basics about forecasting competitons
  • learn characteristics of energy time series data sets (e.g. including energy consumption, energy prices, wind and solar production, etc.)
  • learn to visualize, report and present results
Relevance

The module is highly relevant for practice, not only in the energy industy. Students acquire skills that are useful in data projects, operations and evaluation.

Module Exam

Zum Modul erfolgt eine modulbezogene Prüfung in Form der Entwicklung eines Prognosemodells (20 % der Note), Ausarbeitung zum Modell (Hausarbeit, 50% der Note) sowie Präsentation (in der Regel: 20-40 Minuten, 30 % der Note).

Usage in different degree programs
  • BWL EaFWahlpflichtbereich1st-3rd Sem, Elective
  • ECMXWahlpflichtbereichME5 Economics1st-3rd Sem, Elective
  • GOEMIKWahlpflichtbereich Bereich Betriebswirtschaftslehre1st-3rd Sem, Elective
  • MuUWahlpflichtbereich IIWahlpflichtbereich II B.: Märkte und Unternehmen aus Marktperspektive1st-3rd Sem, Elective
  • VWLWahlpflichtbereich II1st-3rd Sem, Elective
Elements
Name in diploma supplement
Energy Forecasting Competition
Organisational Unit
Lecturers
SPW
4
Language
English
Cycle
irregular
Participants at most
no limit
Preliminary knowledge

Basics in R or python, basics in data science or statistics.

Abstract

In the first third of the Module the students study the competition design, the forecast evaluation methods, benchmark methods and forecasting principles in general in a lecture. The competition task and the corresponding data sets will be released immediately. In the second part the student construct their own forecasting model for the competition and submit their forecasts. Shortly afterwards the results will be released. In the third part of the students write a report on the prediction methods and present their finding.

Contents
  1. Introduction on forecasting competitions
  2. Competition design and reporting of forecasts
  3. Evaluation metrics
  4. Benchmark methods
  5. Options for improving forecasts
Literature
  • Hong, T., Pinson, P., Fan, S., Zareipour, H., Troccoli, A., & Hyndman, R. J. (2016). Probabilistic energy forecasting: Global energy forecasting competition 2014 and beyond.  International Journal of Forecasting, 32(3), 896-913.
  • Makridakis, S., Spiliotis, E., & Assimakopoulos, V. (2020). The M4 Competition: 100,000 time series and 61 forecasting methods. International Journal of Forecasting, 36(1), 54-74.
  • Further Literature will be mentioned during the lecture.
Teaching concept

Classic lectures + Learning by doing

Die Veranstaltung entspricht einem Vorlesungsanteil von 2 SWS und einem Seminaranteil von 2 SWS.

Participants
Lecture with integrated Seminar: Energy Forecasting Competition (WIWI‑C1160)
Module: Energy Forecasting Competition (WIWI‑M0906)