MODEL SELECTION CRITERION IN SURVIVAL ANALYSIS


KARABEY U. , ATA TUTKUN N.

International Conference on Numerical Analysis and Applied Mathematics (ICNAAM), Rhodes, Greece, 19 - 25 September 2016, vol.1863 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 1863
  • Doi Number: 10.1063/1.4992296
  • City: Rhodes
  • Country: Greece

Abstract

Survival analysis deals with time until occurrence of an event of interest such as death, recurrence of an illness, the failure of an equipment or divorce. There are various survival models with semi-parametric or parametric approaches used in medical, natural or social sciences. The decision on the most appropriate model for the data is an important point of the analysis. In literature Akaike information criteria or Bayesian information criteria are used to select among nested models. In this study, the behavior of these information criterion is discussed for a real data set.