基于灰色预测和 BP 神经网络模型的云南省人口总量预测研究

摘要

The population problem has always been a key issue in the economic growth and social stability ,so scientific and accurate forecast of the population is becoming more and more important .In this paper ,by using the total population of yunnan province in China statistical yearbook data ,simulation the population from 2000 to 2013 by the grey prediction model and BP neural network model ,considering the limitations of single forecasting model ,then using the method of standard de‐viation to weight distribution of single forecasting results ,that is the combined forecasting model .Model results show that the combined forecasting model is superior to the single grey prediction model and BP neural network model ,and it has high precision .Finally the population from 2014 to 2025 is forecasted by combined forecasting model ,the results show that the population will exceed 50 million until 2024 ,and the population growth remain at around 0 .6% .%人口问题一直是关系经济增长和社会稳定的关键问题，因此科学准确的人口预测也越来越重要。论文通过利用历年中国统计年鉴上云南省人口总量数据，分别应用灰色预测和 BP 神经网络模型对2000～2013年人口总量数据进行仿真模拟，考虑单一预测模型的局限性，然后利用标准差法对单一预测仿真结果进行权重分配，建立了组合预测模型。模型结果表明，组合预测模型优于单一的灰色预测模型和 BP 神经网络模型，具有较高的精度。最后利用组合预测模型，对2014～2025年的人口总量进行了短期预测，结果表明，到2024年人口总量将突破5000万人，人口增长速度大概维持在0．6％左右。