中国老龄人口死亡率建模及预测

Modeling and Forecasting for Chinese Aging Population Mortality

我国老龄人口死亡率在时间上存在高度线性相关性,并且其对数死亡率的改善速度存在非线性特征.根据我国老龄人口实际死亡率特点,提出一个新的模型,使得这些性质在模型中得到体现.并针对我国老龄人口死亡率数据较少的特点,使用MCMC方法对模型参数进行估计,预测时充分利用参数的分布信息,给出未来几年死亡率的均值和分布情况,从而使预测结果更加可信.

There is a high linear correlation for Chinese aging population mortality about time.It exits nonlinear characteristics to the rate of improvement of logarithm mortality.In this paper,we propose a new model to fit the characteristics of the actual mortality rate of aging population.MCMC method is used to estimate the parameters of the model,which can solve the problem of less data of aging population.Using MCMC methods,the mean and distribution of forecasting mortality are given.The prediction results are more credible because our method utilizes distribution information about the parameters.