PROSPECTIVE MORTALITY TABLE USING THE LEE-CARTER METHOD BASED ON HISTORICAL MORTALITY TABLE

• Main Author: Dini Utami, Meilinda
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/54813
• Institution: Institut Teknologi Bandung

Mortality tables are used for life insurance calculations, such as calculating premiums and reserves. Each year the values on the mortality table will change, this change is because the chance of a person's death each year is uncertain. The result of changes in the values in the mortality table will affect the calculation of life insurance because the calculation requires the values in the mortality table such as ???????? and ????????. In this thesis, the author will predict the probability values in the mortality table using the Lee-Carter method and see their effect on the calculation of premium reserves. This method is implemented in USA mortality data from 2007-2018 with 2007-2016 data used as training data and 2017-2018 as validation data. The simulation results show that the best forecasting result is by using the mortality table data for the past 9 years, with the error size ????????,2017 is 0.001812 and ????????,2018 is 0.003667 while the amount of premium reserves calculated using the forecasted mortality table is smaller than using the existing mortality table.