Mixed-Effects Cox Models for Complex Samples

Mixed-effect proportional hazards models for multistage stratified, cluster-sampled, unequally weighted survey samples. Provides variance estimation by Taylor series linearisation or replicate weights.

svycoxme

The goal of svycoxme is to fit mixed-effects proportional hazards models to data from complex samples. Most of the work is done by the coxme package. The svycoxme package provides wrappers to fit models using survey designs from the survey package, and provides variances estimation by Taylor series linearisation or replicate weights.

Installation

You can install svycoxme from CRAN with:

install.packages("svycoxme")

You can install the development version of svycoxme from GitHub with:

# install.packages("devtools")
devtools::install_github("bdrayton/svycoxme")

Example

This is a basic example using the samp_srcs dataset provided with the package.

library(survey)
library(svycoxme)
des <- svydesign(ids = ~group_id, weights = ~weight, data = samp_srcs)

fit1 <- svycoxme(Surv(stat_time, stat) ~ X1 + X2 + X3 + (1 | group_id), design = des)

summary(fit1)