|
Abstract:
|
The semi-parametric accelerated failure time (AFT) model is a useful alternative to the most frequently used Cox Proportional Hazard (Cox PH) model. It directly links the logarithm of the failure time to the covariates, which offers a nice interpretation. Nevertheless, model checking procedures for the semi-parametric AFT model have been less-developed. In this paper, we illustrate an R package afttest that performs lately proposed model checking methods for the AFT model in terms of the rank-based approach. It provides functions used to verify whether the observed data fit the specific model assumptions such as a functional form of each covariate, a link function, and an omnibus test. The p-value offered in this package is based on the Kolmogorov-type supremum test and the variance of the proposed test statistics is estimated through the re-sampling method. Furthermore, a graphical technique to compare the shape of the observed residual to several the approximated realizations is provided. Simulation studies and illustrations with a well-known Mayo Clinic study (PBC) and Chronic Granulomatous Disease (CGD) Infection Data are described.
|
