What question did this study set out to answer?

To develop a computationally scalable survival modeling approach that performs simultaneous variable selection on both fixed and random effects in high-dimensional clustered settings.

March 10, 2025Statistics in MedicineOpen Access

Efficient Computation of High‐Dimensional Penalized Piecewise Constant Hazard Random Effects Models

View Full Paper

Ask AI

Bookmark

Authors

HHHillary HeilingHarvard University NRNaim U. RashidUniversity of North Carolina at Chapel Hill QLQuefeng LiUniversity of North Carolina at Chapel Hill

Discussion

Member takes

Implication

Methodological study reveals an efficient algorithm for simultaneous variable selection in survival models, suggesting improved biomarker discovery in high-dimensional genomic datasets.

Key Points

To develop a computationally scalable survival modeling approach that performs simultaneous variable selection on both fixed and random effects in high-dimensional clustered settings.
Approximated proportional hazards mixed effects models using a piecewise constant hazard mixed effects survival framework.
Estimated parameters using a modified Monte Carlo expectation conditional minimization (MCECM) algorithm combined with a factor model decomposition of random effects.
Evaluated computational efficiency via numerical simulations and applied the algorithm to a multi-study pancreatic ductal adenocarcinoma gene expression dataset.
Simulation analyses demonstrated that the framework scales effectively to high dimensions and reliably conducts variable selection on fixed and random effects.
Application to pancreatic ductal adenocarcinoma datasets successfully identified key gene expression features linked to patient survival across clustered study cohorts.

Cite This Study

Heiling et al. (2025) studied this question.

synapsesocial.com/papers/6a0fc37e2badbc352afead68 https://doi.org/10.1002/sim.10311

View Full Paper

Ask AI

Bookmark

Implication

Methodological study reveals an efficient algorithm for simultaneous variable selection in survival models, suggesting improved biomarker discovery in high-dimensional genomic datasets.

Efficient Computation of High‐Dimensional Penalized Piecewise Constant Hazard Random Effects Models

Discussion

Member takes

Implication

Key Points

Cite This Study

Efficient Computation of High‐Dimensional Penalized Piecewise Constant Hazard Random Effects Models

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider