A Simulation Study On Kaplan Meier Non-Parametric Survival Methods

Authors

  • K. A. Adeleke DEPARTMENT OF MATHEMATICS, OBAFEMI AWOLOWO UNIVERSITY, ILE - IFE, NIGERIA

Abstract

Exploring a time to event data, especially time to failure (Death) assuming some data are censored and no tied observations. This article discusses the use of Kaplan Meier nonparametric approach on mortality data simulated over some period specifically 15-weeks follow up. Series of simulations were carried out and survival and hazards probabilities and Kaplan Meier graphs were obtained for different simulations. Law of large numbers as well as demographic stochasticity in the model were observed. The findings suggested that as the number of subjects at risk increases, the expected survival µbwhich is approximately 6 weeks and unconditional probability of survival S(t) are almost decreasing at the same rate indicating more mortality on daily basis while the hazards h(t) are tending towards being constant.

Downloads

Published

2021-04-12

How to Cite

Adeleke, K. A. (2021). A Simulation Study On Kaplan Meier Non-Parametric Survival Methods. Journal of the Nigerian Mathematical Society, 31(1-3), 243–254. Retrieved from https://ojs.ictp.it/jnms/index.php/jnms/article/view/765

Issue

Section

Articles