Nonlinear mixed-effects state space models with applications to HIV dynamics

摘要

Nonlinear state space models with mixed-effect (NLMESSM) are proposed to model HIV clinical longitudinal data. With NLMESSM, filtering algorithms are proposed to estimate the individual/population states. Maximum likelihood via iterated filtering and variance components model are proposed to estimate fixed/random effects respectively. Simulation results validate the effectiveness of NLMESSM.