【摘要】 本文提出并证明了平均预期寿命增量与对应生命表中尚存人数增量成等比例关系的命题；给出了利用命题编制我国省级地区研究期内各年生命表的方法、以及基于编制的生命表预测老年人数峰值期的步骤；介绍了构建拟合样本精度高且老年人数预测值满足峰值期的向量自回归模型的方法。最后以上海市为例进行实证分析，结果表明：用文中介绍的方法建立模型并进行预测，有助于提高总人数、老年人数以及老龄化系数中长期预测值的精度。

关键词  老龄化系数  生命表  峰值期  VAR模型

中图分类号  C921      文献标识码  A

Research on the Middle and Long Term Forecast Method of the Aging Coefficient of Provincial Regions in China

Lv Shengge

College of Economics and Trade of Hangzhou Dianzi University

Abstract: This article proposes and proves a proportional relationship between the increment of average life expectancy and the number of people who are still alive in the corresponding life table, gives the method of establishing complete life tables of each year of China's provincial regions during the research period by proposition compilation, describes the steps of predicting the peak period of aging population based on compiled life tables, and constructs a fitting sample with a high accuracy and the predictive value of aging people consistent with the Vector Autoregression Model method. Finally, this article conducts an empirical analysis by taking Shanghai as the example and the results show that it helps increase the overall number of people and aging people and the accuracy of the middle and long term predictive value of aging coefficient through modeling and prediction by using the method introduced in this article.

Key words: Aging Coefficient; Life Table; Peak Period; VARModel