Estimating survival probabilities for pensioners of life insurance corporation (LIC) of India

Survival Analysis encompasses a wide variety of methods for analyzing the timing of events. The term Survival Analysis pertains to a statistical approach designed to take into account the amount of time an experimental unit contributes to a study. It is the study of time between entry of the observation and subsequent event. Originally the event of interest was death and hence the term Survival Analysis. The wheel of Survival Analysis has been reinvented several times in different disciplines where terminology varies from discipline to discipline. Survival Analysis finds a variety of applications in social and economic sciences. In the insurance sector these models are used to study the population survival pattern to make use of them in determining the premium rates applicable to people at different ages.

The present study aims at analyzing the survival patterns for the sample of policyholders of Lice Insurance Corporation (LIC) of India with specific reference to pension plan: New Jeevan Suraksha –I. using various survival models. The three major approaches: non parametric, semi parametric, and parametric approach for the estimation of the survival models are applied to the sample data and the results are compared. We have carried out Survival Analysis using different models such as Cox-Regression model, General Linear model, Life Table method, Kaplan-Meier method and survival probabilities are computed in each of the cases. The study also helps identifying the major factors affecting the hazard rate for Indian population.