EcoSta 2022: Start Registration
View Submission - EcoSta2022
A0366
Title: Optimal subsampling for massive survival data Authors:  HaiYing Wang - University of Connecticut (United States) [presenting]
Abstract: With increasingly available on massive survival data, researchers need valid and computationally scalable statistical methods for survival modeling. Existing works focus on relative risk models using the online updating and divide-and-conquer strategies. The focus is on using optimal subsampling algorithms to tackle the computational issues in analyzing survival data. We first discuss the results on fast approximation to the maximum likelihood estimator for a parametric Weibull accelerate failure time model, and then present our current state of knowledge and the challenges of optimal subsampling in the context of semiparametric models with censored data.