EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Gradient boosting-based discrete failure time model for selecting time-varying effects and interactions

Lingfeng Luo, Kevin He, Jeremy M G Taylor

Lifetime Data Analysis · 2026-06 · 32 (2)

DOI: 10.1007/s10985-026-09710-4 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Lifetime Data Analysis, online erschienen am 13.05.2026. Verfasst von Lingfeng Luo und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biostatistics, University of Michigan, Ann Arbor, MI, USA. Hauptthemen laut MEDLINE: Models, Statistical. Die Arbeit nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIH HHS, National Institutes of Health.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Analyzing survival data from the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program offers invaluable insight to guide cancer management. In these data the event times are recorded in units of months. In such datasets, it can be important to consider the selection of covariates, distinguish between time-varying and time-independent effects, and capture important interaction terms. However, variable selection is difficult to combine with the differentiation between time-varying and time-independent effects. Moreover, accurately selecting interaction terms is more challenging because of the hierarchy restriction that an interaction term should only enter a model if one or both associated main effects are also included. To resolve these issues, we develop a coordinate ascent-based gradient boosting procedure for discrete failure time models. This approach achieves variable selection in high-dimensional settings, differentiates between time-varying and time-independent variables, and selects important interaction terms under hierarchy restrictions. Traditional boosting stopping criteria are ad hoc, while preferable information-criteria-based rules necessitate defining degrees of freedom, a challenge in survival models. Our approach provides well-defined degrees of freedom and offers information-criteria-based stopping rules. Simulation studies show the proposed method achieves good selection performance. We apply the proposed method to SEER melanoma cancer data and select important risk factors, distinguish between time-varying and time-dependent effects and detect important two-way interactions.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Models, StatisticalSEER ProgramHumansBoosting Machine Learning AlgorithmsSurvival AnalysisComputer SimulationTime Factors

Stichwörter der Autor:innen

Discrete failure timeHierarchical interactionsHigh dimensional variable selectionSurvival analysisTime-varying effects

Autor:innen und Institutionen

  • Department of Biostatistics, University of Michigan, Ann Arbor, MI, USA.
  • Department of Biostatistics, University of Michigan, Ann Arbor, MI, USA. jmgt@umich.edu.

Förderung

  • NIH HHS
  • National Institutes of Health

Zitieren

APA

Luo, L., He, K., & Taylor, J. M. G. (2026). Gradient boosting-based discrete failure time model for selecting time-varying effects and interactions. Lifetime Data Analysis, 32(2). https://doi.org/10.1007/s10985-026-09710-4

Vancouver

Luo L, He K, Taylor JMG. Gradient boosting-based discrete failure time model for selecting time-varying effects and interactions. Lifetime Data Analysis. 2026;32(2). doi:10.1007/s10985-026-09710-4

BibTeX
@article{Luo2026,
  author = {Luo, Lingfeng and He, Kevin and Taylor, Jeremy M G},
  title = {Gradient boosting-based discrete failure time model for selecting time-varying effects and interactions},
  journal = {Lifetime Data Analysis},
  year = {2026},
  volume = {32},
  number = {2},
  doi = {10.1007/s10985-026-09710-4}
}
Kostenlos starten →