EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Predictive Modeling of Retirement Decisions Based on Health Data: An AI-Powered Analysis

Tahira Waseem, Mahnoor Ghalib

Journal of Computational Science and Applications · 11.12.2025 · 2 (2) · 31-37

DOI: 10.51846/jcsa.v2i2.4554 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Computational Science and Applications, online erschienen am 11.12.2025. Verfasst von Tahira Waseem und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Retirement choices are a complicated interaction of health, socioeconomic status, and pension schemes. These outcomes have been studied traditionally using econometric and actuarial models that are linear nonlinearities and high dimensional dependencies in health and social data can often not be represented in assumptions. This article uses artificial intelligence (AI) techniques to enhance predicting the retirement decision based on the data provided in the Health and Retirement Study (HRS), is a massive longitudinal survey of adults of the US aged 50 and above. We benchmark traditional, logistic regression and Cox proportional hazards versus Random Forests, XGBoost, and Neural Networks. The cross-validation of 10-fold competence demonstrates that AI designs, especially XGBoost and Neural Networks, perform better have significantly greater predictive accuracy (AUC to 0.85) compared with econometric baselines. Explainability analyses SHAP and LIME are used to identify burden of chronic illness, functional limitations, hospitalizations history, and self-rated health as its predominant predictors, and pension eligibility and income. This proves that AI has the capability to capture nonlinear relations between health- and socioeconomic determinants, which provide interpretable and policy-relevant information to plan retirement, pension sustainability as well as preventive health interventions.

Zitieren

APA

Waseem, T., & Ghalib, M. (2025). Predictive Modeling of Retirement Decisions Based on Health Data: An AI-Powered Analysis. Journal of Computational Science and Applications, 2(2), 31-37. https://doi.org/10.51846/jcsa.v2i2.4554

Vancouver

Waseem T, Ghalib M. Predictive Modeling of Retirement Decisions Based on Health Data: An AI-Powered Analysis. Journal of Computational Science and Applications. 2025;2(2):31-37. doi:10.51846/jcsa.v2i2.4554

BibTeX
@article{Waseem2025,
  author = {Waseem, Tahira and Ghalib, Mahnoor},
  title = {Predictive Modeling of Retirement Decisions Based on Health Data: An AI-Powered Analysis},
  journal = {Journal of Computational Science and Applications},
  year = {2025},
  volume = {2},
  number = {2},
  pages = {31--37},
  doi = {10.51846/jcsa.v2i2.4554}
}
Kostenlos starten →