A0822
Title: Comparison of the survival analysis with functional convex clustering and joint model for complex data
Authors: Yuko Araki - Tohoku University (Japan) [presenting]
Abstract: Two methods are considered for cases where the pattern of change in covariates varies over time, and space affects survival time. One is a Cox proportional hazards model incorporating functional convex clustering, and the other is a joint model. The former facilitates capturing spatiotemporal variations and naturally incorporates the patterns of dynamics as covariates in survival time analysis. On the other hand, the joint model, by handling both longitudinal data and survival time simultaneously, is known to mitigate bias in survival analysis results due to temporal changes in covariates. The behavior of the two models is examined and compared through numerical experiments.
