Lee-Carter死亡率模型的估计与应用——基于中国人口数据的分析

文章选择Lee-Carter死亡率模型对中国人口死亡率数据进行拟合和预测,以探讨Lee-Carter模型在中国的适用性和表现形式。基于1992～2007年中国人口分年龄组死亡率数据,文章对奇异值分解法(SVD)、最小二乘法(OLS)、加权最小二乘法(WLS)和极大似然法(MLE)的拟合结果和预测能力进行了比较分析。结果表明,加权最小二乘法具有最好的拟合和预测效果。文章利用最优的拟合模型,对未来中国人口平均预期寿命进行了预测,并利用Bootstrap方法进行了区间估计。

Estimation and Application of the Lee-Carter Model:Based on Demographic Data of China

The main objective of this paper is to study the performance and application of the LeeCarter Model in China.Using Chinese agespecific mortality data from 1992 to 2007,this paper estimates the LeeCarter Model by the methods of singular value decomposition,ordinary least square,weighted least square and maximum likelihood estimation.The empirical results indicate that the weighted least square method provides the best performance,with respect to the goodnessoffit and predictive ability.Based on the optimally...