Conditional mix-GEE models for longitudinal data with unspecified random-effects distributions

Semiparametric inference for proportional mean residual life models with clustered survival data

“…Based on the penalized QIFs approach, the assumption of an explicit distribution for random effects was completely relaxed and both the serial correlation and heterogeneous variation arose in longitudinal data were addressed. Other applications of conditional inference can be found in recent literature, including Cho et al (2017) and Xing et al (2018) for improving the efficiency of estimators, Yu et al (2018) for establishing asymptotic properties of estimators under misspecified GLMMs and Liu and Xiang (2019) for the inference of GLMMs with the presence of missing covariates and without distributional assumption of random effects.…”