| afttest-package | R Documentation |
The afttest package provides diagnostic tools for semiparametric accelerated failure time (AFT) models. It supports formal model checking and graphical assessment for fitted AFT models, providing a unified framework for evaluating model adequacy and interpreting results.
The main interface is the generic function afttest, which
accepts either a formula or fitted model objects from the aftgee
package, including aftsrr and aftgee.
Implemented diagnostic procedures include:
Omnibus test (testType = "omnibus"), for assessing overall model adequacy;
Link function test (testType = "link"), for evaluating the specified link function;
Functional form test (testType = "covForm"), for checking whether a continuous covariate is correctly specified.
The package supports diagnostic inference for:
rank-based estimators with non-smooth (eqType = "ns") and induced-smoothed (eqType = "is") estimating equations;
least-squares estimators (estMethod = "ls") fit through generalized estimating equations.
For computational efficiency, afttest implements a fast multiplier
bootstrap procedure. When linApprox = TRUE, an asymptotic linear
approximation is used to avoid repeated iterative model fitting, substantially
reducing computation time relative to standard resampling with
linApprox = FALSE.
Maintainer: Woojung Bae matt.woojung@gmail.com (ORCID)
Authors:
Bae, W., Choi, D., Yan, J., & Kang, S. (2025). afttest: Model diagnostics for semiparametric accelerated failure time models in R. arXiv preprint arXiv:2511.09823.
Choi, D., Bae, W., Yan, J., & Kang, S. (2024). A general model-checking procedure for semiparametric accelerated failure time models. Statistics and Computing, 34(3), 117.
Useful links:
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.