径向基神经网络模型在人口老龄化预测中的应用——以湖南省为例

Radial Basis Function Neural Network Model Applied in the Forecast of Population Aging——Taking Hunan Province as an Example

人工神经网络具有良好的非线性映射逼近性能,在各类预测研究中得到了广泛应用。径向基函数神经网络(Radial Basis Function,RBF)因其网络学习速度较快且能避免局部极小值,预测值则更接近于真实值。针对湖南省老龄化突出的现状,以湖南省老龄化指数历史数据为基础,从经济水平、人口自然增长、社会保障三个方面构建湖南省老龄化的影响因子体系,用RBF神经网络方法建立了人口老龄化的定量预测模型。作为对比,同时用多元线性回归方法进行了预测,结果表明RBF神经网络预测模型精度更高,预测结果更加合理可靠。

Artificial neural network has a good nonlinear mapping approximation performance, and it has been widely applied in all kinds of prediction. Radial Basis Function （RBF） is comparatively fast in network learning speed and able to avoid local minima, so its predictive value is more close to the true one. Aiming at the outstanding of Hunan population aging, based on the aging index historical data, this paper constructed an impact factor system from three aspects such as economic level, the natural population growth and social security. It also constructed a quantitative population aging prediction model by RBF neural network model. In contrast, this paper adopted multiple linear regression method to predict too. The result showed that the RBF neural network model was more accurate and reliable.