VIRTUAL EVENT: The link between previous life trajectories and a later life outcome
27 January 2021, 1:00 pm–2:00 pm
In this webinar, Matthias Studer will discuss how to investigate the link between previous life trajectories and a later life outcome using a feature selection approach.
This event is free.
Event Information
Open to
- All
Availability
- Yes
Cost
- Free
Organiser
-
Bozena Wielgoszewska
Several studies have investigated the link between a previous trajectory and a given outcome later on in life.
However, trajectories are complex and it is not straightforward to identify which aspects of the trajectories are relevant in terms of both prediction and testing specific theories.
In this talk, Matthias will discuss work conducted with Danilo Bolano, in which they propose an innovative approach based on sequence analysis and data mining feature selection algorithms.
Furthermore, Matthias will discuss the use of two feature selection approaches, namely random forest (Boruta) and a LASSO (Stability Selection).
The proposed approach is illustrated through a study that explores the effects of family and work trajectories between ages 20 and 40 on women’s earnings in midlife.
QSS seminar series
In this weekly Quantitative Social Science (QSS) seminar series, speakers present research that falls under the broad umbrella of quantitative social science.
Links
About the Speaker
Matthias Studer
Associate Professor at The Swiss National Centre of Competence in Research LIVES (NCCR LIVES) and the Institute of Demography and Socioeconomics, University of Geneva
Matthias is a recognised specialist in the quantitative analysis of longitudinal data and more specifically in sequence analysis. Since December, he has been President of the Sequence Analysis Association.
More about Matthias StuderOther events in this series