A comparative simulation study for estimating accelerated failure time models

Sangbum Choi
2018 Journal of the Korean Data and Information Science Society  
Semiparametric accelerated failure time (AFT) models directly relate the predicted failure times to covariates and are a useful alternative to Cox's proportional hazards models that work on the hazard function or the survival function. In this paper, we briefly review different approaches to estimate the AFT model and evaluate their performance with finite samples via extensive simulation studies. Specifically, we compared (i) inverse probability of censoring weighted (IPCW) least squares, (ii)
more » ... log-rank estimator, (iii) Gehan-type log-rank estimator, (iv) Buckley-James estimator, and (v) nonparametric maximum likelihood estimator (NPMLE). Overall, rank-based estimators and Buckley-James estimator are efficient and relatively more robust to distributions of residual and censoring variables, whereas the IPCW estimator is very sensitive to distribution and amount of censoring. The NPMLE is asymptotically efficient and useful as it allows for hazard-based formulation, and thus can be used to analyze more structured survival data.
doi:10.7465/jkdi.2018.29.6.1457 fatcat:dnu5porx7rattalbyzz3a2ai24