摘要
由于高龄人口样本数据少,死亡率计算结果可信度低,传统的基于年龄的死亡率外推方法容易产生偏差。本文采用Age-Shifting模型,选取国家统计局和我国寿险业经验生命表的人口死亡率数据,考虑时间和年龄两个维度,对高龄人口死亡率进行动态拟合,并将拟合结果与传统的Gompertz模型进行比较。研究表明,Age-Shifting模型在拟合高龄人口死亡率的过程中,不仅考虑了年龄上的拟合度,同时也考虑到死亡率随时间递减的趋势,克服了Gompertz模型的缺陷,使得高龄人口死亡率的拟合结果更科学、合理。
Because of the limited statistics for elderly population samples, mortality estimates have a low credibility and the traditional age-based mortality extrapolation methods are prone to bias. In this paper, the Age-Shifting model was adopted and the mortality data from the NBS and life tables of the life insurance sector was selected to dynamically fit the mortality of the elderly population, taking into consideration the two dimensions of time and age. Then the results were compared with the traditional Gompertz model. The results showed that the Age-Shifting mod- el not only took into account the fitting degree of age, hut also the tendency of death rate decreasing with time in fit- ting the mortality rate of the elderly population. Therefore, the Age-Shifting method overcame the shortcomings of the Gompertz model to make the fitting result of the elderly population mortality more scientific and reasonable.
作者
赵明
ZHAO Ming(Postdoctoral Workstations, China HUARONG Asset Management Co. , Ltd. , Beijing 100033 Postdoctoral Research Station, School of Economics and Management, Wuhan University, Wuhan 430072)
出处
《保险研究》
CSSCI
北大核心
2017年第1期37-45,共9页
Insurance Studies
基金
国家社科基金重大项目(13&ZD164)
国家自然科学基金项目(71173230)的资助
