ANALYSIS NON-PARAMETRIC SURVIVAL MODEL WITH KAPLAN-MEIER METHOD FOR LIFE INSURANCE DATA

• Main Author: Riani, Winda
• Format: Final Project
• Language: Indonesia
• Subjects: Matematika
• Online Access: https://digilib.itb.ac.id/gdl/view/34195
• Institution: Institut Teknologi Bandung

Survival Analysis is a specialized statistical analysis is used to analyze data that the variables relating to the time until the appearance of an event. In this research, survival analysis was used to analyze the time until someone dies, because in life insurance is important to know a person's chance of dying at a given time. Non-parametric basis, the Kaplan-Meier method is a popular method used for sensor data, and can provide a definite proportion of survival because it uses the exact survival time, not based on class intervals. The main purpose of this final project was to study and analyze the estimated survival function of real data in the Canadian life insurance companies that consisting of a pair of men and women policyholders. With this method we can estimate survival function, which can be presented in graphs and table mortality. For interval estimation of the survival function we use the greenwood’s formula, with results that are small enough for this data. The test results obtained using Log-rank test is survival function of men and women are not the same. Men have a greater hazard rate than women.