CMStatistics 2022: Start Registration
View Submission - CMStatistics
B0383
Title: Modelling non-homogeneous censored time-to-event data using semiparametric accelerated failure time model Authors:  Iketle Maharela - University of Pretoria (South Africa) [presenting]
Din Chen - University of Pretoria (South Africa)
Lizelle Fletcher - University of Pretoria (South Africa)
Abstract: In survival analysis, classical accelerated failure time (AFT) models provide a useful alternative to the usual proportional hazards models in analysing the associations between covariates and time-to-event data. In most cases, the basic assumption is that the event time data being analysed are homogeneous in that some covariates influencing the hazard function for an individual may not be observed or measured. Semi-parametric AFT model has recently been developed to analyse both homogeneous and heterogeneous survival data. We illustrate the performances of these models with an extensive simulation study and with application to real datasets.