Interval Censored Multi-State Models

Allows for the non-parametric estimation of transition intensities in interval-censored multi-state models using the approach of Gomon and Putter (2024) or Gu et al. (2023) .

icmstate

Interval-Censored Multi-STATE modelling (icmstate)

Using the package it is possible to non-parametrically estimate transition intensities in interval-censored multi-state models. The package also provides utilities for plotting panel data and transition probabilities between states. Many of the functions from the package can be used for visualisation as well. The package currently supports two approaches to determine transition intensities, either using the Multinomial likelihood approach as in Gomon and Putter (2024) or the latent Poisson approach as in Gu et al (2023).

Using the package

More information on how to use the package can be found in the vignettes under the doc directory.

Installation

Install the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("d-gomon/icmstate")

or the released version from CRAN using:

install.packages("icmstate")

Research Code

The results of Gomon and Putter (2024) can be reproduced using the code in the folder Research_Code.

References

Gomon D., Putter H. (2024): Non-parametric estimation of transition intensities in interval censored Markov multi-state models without loops, arXiv

Gu Y., Zeng D., Heiss G., Lin D. Y. (2023): Maximum likelihood estimation for semiparametric regression models with interval-censored multistate data, Biometrika