A1033
Title: A functional joint model for longitudinal and survival data with a baseline functional covariate
Authors: Luo Xiao - North Carolina State University (United States) [presenting]
Abstract: Motivated by the Women's Health Initiative (WHI) OPACH ancillary study, a joint model for longitudinal and survival data is proposed with a baseline functional covariate. The longitudinal data is modeled by functional principal components and links to the survival data via functional principal component scores. In both sub-models, scalar-on-function regression forms are used to account for the effect of the baseline functional covariate. For model estimation, an efficient EM algorithm coupled with penalized splines is proposed. The proposed model gives sensible associations of accelerometry-data-derived sitting-bout-length profiles of older women in the OPACH study with physical function and all-cause mortality.
