Danilo Bolano , University of Florence
Matthias Studer, Université de Genève and NCCR-LIVES
This article proposes an innovative approach based on data mining feature selection algorithms to study the relationships between a previous trajectory and a later-life outcome. The procedure works in two steps. It starts by automatically extracting key properties of the previous trajectory coded as sequences. These properties aims to capture the key life-course aspects of trajectories, namely sequencing, timing, and duration. Second, it uses feature selection algorithms to identify the most relevant properties associated with the outcome. The proposed approach is illustrated through a study of the effects of family and work trajectories on women’s income in midlife.
Presented in Session 48. Data and Methods