Humboldt-Universität zu Berlin - School of Business and Economics

Applied Predictive Analytics

The model give students an opportunity to participate in a real-world forecasting challenge related to planning problems  in business areas such as marketing, finance, or others. In this scope, students have the opportunity to develop a variety of skills, including:

 

  • Working in a real-world project setting allows students to further develop their team work and project management abilities.
  • Students get acquainted with contemporary software packages for predict analytics.
  • Students are able to develop advanced forecasting models using a variety of techniques from statistics, machine learning, and other domains.
  • Students advance their knowledge in data integration, preparation, and transformation which allows them to create predictive variables from noisy real-world data sets.

 

Topic & Content

  • The module involves participating in a real-world forecasting competition such as the annual data mining cup, the ACM KDD cup, or a kaggle challenge. In this scope, students will experience several typical challenges that arise in real-world modeling projects, and develop the necessary skills to overcome these obstacles.
  • Development of a competition entry (typically a prediction model) for a specified forecasting challenge (50%), studying relevant literature (25%), preparation of a seminar presentations (25%)