Estimation of Survival Function

In the past decade applications of the statistical methods for survival data analysis have been extended beyond biomedical and reliability research to other fields. The term survival data has been used in a broad sense for data involving time to a certain event such a failure, response, death and so on. Survival times are subjected random variations and like any random variable, they form a distribution. The ability to estimate a survival distribution in the presence of censoring is important and has been studied extensively. This paper is concerned with estimators of survival function. If one is not willing to make parametric assumptions about the exact form of the underlying survival and censoring distributions but is willing to assume independence between survival and censoring variables, Kaplan and Meier provided an estimator which is consistent, among other desirable properties.