A semiparametric nonlinear mixed-effects model with non-ignorable missing data and measurement errors for HIV viral data

Abstract

Semiparametric nonlinear mixed-effects (NLME) models are very flexible in modeling long-term HIV viral dynamics. In practice, statistical analyses are often complicated due to measurement errors and missing data in covariates and non-ignorable missing data in the responses. We consider likelihood methods which simultaneously address measurement error and missing data problems. A real dataset is analyzed in detail, and a simulation study is conducted to evaluate the methods.