Using machine learning to uncover the relation between age and life satisfaction

Micha Kaiser; Steffen Otterbach; Alfonso Sousa-Poza

doi:10.1038/s41598-022-09018-x

Using machine learning to uncover the relation between age and life satisfaction

Sci Rep. 2022 Mar 28;12(1):5263. doi: 10.1038/s41598-022-09018-x.

Authors

Micha Kaiser¹, Steffen Otterbach^{2

3}, Alfonso Sousa-Poza^{2

3}

Affiliations

¹ Department of Management, Society, and Communication, Copenhagen Business School, Dalgas Have 15, 2000, Frederiksberg, Denmark. mka.msc@cbs.dk.
² Institute for Health Care and Public Management, University of Hohenheim, Fruwirthstrasse 48, 70599, Stuttgart, Germany.
³ IZA, Bonn, Germany.

Abstract

This study applies a machine learning (ML) approach to around 400,000 observations from the German Socio-Economic Panel to assess the relation between life satisfaction and age. We show that with our ML-based approach it is possible to isolate the effect of age on life satisfaction across the lifecycle without explicitly parameterizing the complex relationship between age and other covariates-this complex relation is taken into account by a feedforward neural network. Our results show a clear U-shape relation between age and life satisfaction across the lifespan, with a minimum at around 50 years of age.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Longevity*
Machine Learning
Personal Satisfaction*