Pseudo Maximum Likelihood Estimation for Semi-Competing Risks Data.

PMLE4SCR

This R package implements two-stage pseudo maximum likelihood estimation (PMLE) and simultaneous maximum likelihood estimation (MLE) for copula-based regression models with semi-competing risks data. The marginal distributions are modeled by semiparametric transformation regression models, and the dependence between the bivariate event times is specified by a parametric copula function with the copula parameter that may depend on some covariates.

Install PMLE4SCR

To install this R package:

install.packages("PMLE4SCR")

Example

library(PMLE4SCR)
data(BMT, package = "SemiCompRisks")
BMT$g <- factor(BMT$g, levels = c(2, 3, 1),
                labels = c("AML-low", "AML-high", "ALL"))

# Two-stage PMLE
myfit <- PMLE4SCR(BMT, time = "T2", death = "T1",
                  status_time = "delta2", status_death = "delta1",
                  T.fmla = ~ g, D.fmla = ~ g,
                  copula.family = "Clayton",
                  copula.control = list(formula = ~ g))
myfit$gamma
myfit$betaT

# Simultaneous MLE
myfit_mle <- MLE4SCR(BMT, time = "T2", death = "T1",
                     status_time = "delta2", status_death = "delta1",
                     T.fmla = ~ g, D.fmla = ~ g,
                     copula.family = "Clayton",
                     copula.control = list(formula = ~ g))
myfit_mle$gamma
myfit_mle$betaT

Reference

This package implements the methods described in:

Arachchige, S. J., Chen, X., and Zhou, Q. M. (2025). Two-stage pseudo maximum likelihood estimation of semiparametric copula-based regression models for semi-competing risks data. Lifetime Data Analysis, 31, 52-75.