Predicting employment trajectories and transitions into retirement using machine learning

The aim of the project is to use machine learning methods to generate improved predictions of individual retirement decisions. Data from the statutory pension insurance system provides a good basis for such predictions. In particular, they contain detailed employment biographical information, which is the most important basis for extrapolating individual employment histories. In particular, supervised machine learning methods are to be used for the analysis, which are linked to the duration analysis. In addition, the potential of methods from the field of natural language processing for analysing biographical data will be explored.