欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
由于人口死亡率年龄跨度长，不同年龄段人口的生命特征存在较大差异，参数修匀方法需要针对不同年龄段分别建模，且不能反映人口死亡率随时间改善这一趋势。本文针对我国动态生命表，采取非参数二维修匀方法，选取二维Beta核方法与二维泊松P-样条方法，对全年龄人口死亡率进行修匀，并对两种方法的修匀结果进行比较。研究表明：在全年龄段上，二维Beta核密度修匀法的拟合度高于二维泊松P-样条修匀法，但在边端年龄处和高龄人口年龄段，两种方法的修匀结果各有优劣，将两种方法有机结合，能够提高生命表的修匀质量。
英文摘要：
Because of the long age span of population mortality, there are significant differences in the life charac- teristics of the population at different ages. The smoothing method of parameters should be modeled for different age groups, and it can＇ t reflect the improving trend of population mortality rates over time. In this paper, the two-dimensional Beta kernel density method and two-dimensional Poisson P-spline method were used to reconstruct the mortality rates of the whole age and smoothed them with the nonparametric smoothing method, and the results of the two methods were compared. The results showed that the fit of the two-dimensional Beta kernel density method was better than the two-dimensional Poisson P-spline method for the whole age. But for the side ages and the age of the elderly population, the results of the two methods had respective advantages and disadvantages. If the two methods are combined, it can improve the smoothing quality of the life table.
同期刊论文项目
同项目期刊论文
期刊信息
位置：