A0211
Title: Semiparametric mixture regression for asynchronous longitudinal data
Authors: Yehua Li - University of California at Riverside (United States) [presenting]
Abstract: In an extensive longitudinal investigation of women's health during mid-life and beyond, known as the Study of Women's Health Across the Nation (SWAN), hormonal biomarkers are repeatedly assessed, following an asynchronous schedule compared to other error-prone covariates, such as physical and cardiovascular measurements. A subgroup analysis of the SWAN data is conducted, employing a semiparametric mixture regression model, which allows the exploration of how the relationship between hormonal responses and other time-varying or time-invariant covariates varies across subgroups. To address the challenges posed by asynchronous scheduling and measurement errors, the time-varying covariate trajectories are modeled as functional data with reduced-rank Karhunen-Loeve expansions, where splines are employed to capture the mean and eigenfunctions. Treating the latent subgroup membership and the functional principal component scores as missing data, an EM algorithm is proposed to effectively fit the joint model, combining the mixture regression for the hormonal response and the FPC model for the asynchronous, time-varying covariates. In addition, data-driven methods are explored to determine the optimal number of subgroups within the population. Through the comprehensive analysis of the SWAN data, a crucial subgroup structure is unveiled within the aging female population, shedding light on important distinctions and patterns among women undergoing menopause.
