Title: Functional survival data analysis with direct and indirect effects for high dimensional data
Authors: Yuko Araki - Shizuoka University (Japan) [presenting]
Abstract: Statistical methods are introduced for survival analysis which contains complex associations of several variables including very high dimensional intermediate variable. The proposed model deal with such high dimensional data as functional data with devices of basis expansions and sparse PCA for dimension reduction. The model is constructed based on structural equation modeling (SEM) for survival outcome. We extend SEM to be able to discuss a causal inference. The crucial issue is how to select the regularization parameters used in model building process. We introduced a model selection criterion to select this value. The proposed models were evaluated through simulations and real data example.