A PARAMETRIC APPROACH ON ACCELERATED FAILURE TIME MODEL

Abstract

Survival analysis techniques have become standard tools for statistician in controlled clinical trial. The application of survival models to clinical trial data is valid when the endpoint of interest is the “time to the occurrence of a particular event”. The Cox’s proportional hazard model and its extension are used comprehensively for the past two decades. The Accelerated failure time(AFT) model is also presented and competing to proportional hazards model in the analysis of clinical trial data where the effect of treatments are to accelerate the event of interest in a tuberculosis controlled clinical trial over a treatment period of six months. It is concluded that the proportional hazards models demonstrate significant lack of fit while the accelerated failure time model in this and an illustration about the added advantages of AFT model