论文部分内容阅读
任何国家在编制生命表时都面临着高年龄段可以观察到的数据量较小的局限。本文首先对我国第五次人口普查数据及日本死亡率数据进行了初步分析,同时对死亡率的基本特点进行了比较,在此基础上,选择80～94岁的死亡率数据作为建模依据;并对不同参数模型进行拟合,选择了拟合度较好的 Logistic2、Logistic3、Kannisto 和 HP模型作为高年龄段死亡率趋势外推模型;最后利用前面讨论的模型对我国寿险业1990～1993和2000～2003生命表进行高年龄段的趋势预测,并比较了不同模型下预期寿命的差异。
When compiling a life table, any country is confronted with a small amount of data that can be observed in higher age groups. In this paper, we first conducted a preliminary analysis of the data of the fifth national census in China and Japanese mortality data and compared the basic characteristics of the mortality rates. Based on this, we selected the mortality data of 80-94 years old as the modeling basis. The Logistic2, Logistic3, Kannisto and HP models with good fitting degree were selected as the trend extrapolation models for the high-age population. Finally, the models discussed above were used to analyze the relationship between China’s life insurance industry from 1990 to 1993 and 2000 ~ 2003 life table to predict the trend of high age, and to compare the different life expectancy under different models.