Modified Weighted Kaplan-Meier Estimator

Abstract

In many medical studies majority of the study subjects do not reach to the event of interest during the study period. In such situations survival probabilities can be estimated for censored observation by Kaplan Meier estimator. However in case of heavy censoring these estimates are biased and over estimate the survival probabilities. For heavy censoring a new method was proposed (Bahrawar Jan, 2005) to estimate the survival probabilities by weighting the censored observations by non-censoring rate. But the main defect in this weighted method is that it gives zero weight to the last censored observation. To over come this difficulty a new weight is proposed which also gives a non-zero weight to the last censored observation.