Constructing Life Tables by Marital status in Taiwan

• Main Author: 蔡乙瑄
• Other Authors: 余清祥
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/36125088373671285671

碩士 === 國立政治大學 === 風險管理與保險研究所 === 101 === Marriage is one of important institutions in the human society. People think that the couple can obtain a stable life, good physical and mental condition through division of labor, in addition to living longer. Many studies showed that the married mortality is lower than those of other marital status, but most of them used the sample data rather than population data. In this study, we use Taiwan population data provided by Ministry of the Interior to explore if the married have lower mortality rates. The data used are of the period 1994~2011, in the format of 5-age group between ages 0 and 99), and are divided according to three marital status: single, married, and divorced/widowed. We shall first construct the life table by marital status. Then, we evaluate if it is feasible to consider the marital status as a risk factor in pricing life insurance products. Furthermore, we use the Lee-Carter model to calculate the mortality improvement for various marital status. From Taiwan’s mortality data, we found that, given 3% of interest rate and comparing to the whole population, the married male and female have 4%~8% and 4% of discounts for whole life insurance, and 9%~41% and 14%~34% of discounts for term life insurance. Applying the Lee-Carter model, we found that the married in general have better mortality improvement than the whole Taiwan population. If these trends of mortality improvements continue, the gap of mortality rates between the married and other marital status will be bigger in the future. In other words, the marital status can be chosen as a risk factor for modeling life insurance products.