ARIMA模型在中国城镇人口预测中的研究  

Research on ARIMA Model in Urban population Forecast in China

作　　者：沈进 许玲[1] SHEN Jin;XU Ling(Anhui Sanlian University,Hefei 230601,China)

机构地区：[1]安徽三联学院,安徽合肥230601

出　　处：《遵义师范学院学报》2025年第2期75-78,共4页Journal of Zunyi Normal University

基　　金：安徽省高校自然科学重点项目(2024AH050507);安徽省高校自然科学重点项目(2023AH051708)。

摘　　要：根据国家统计局1970-2023年的中国城镇人口数据,创建了基于ARIMA算法的时间序列模型,对中国城镇人口进行预测。根据模型定阶和参数估计确定参数范围,再将预测结果与实际值进行比较,得出ARIMA(1,1,0)模型对中国城镇人口的预测最为准确。最后将ARIMA(1,1,0)模型与指数曲线拟合模型进行对比,发现ARIMA模型在人口预测上具有更高的精确度。该模型显示,中国城镇人口在短期内会继续增长,但增长趋势会逐渐变缓。According to the urban population data of China from 1970 to 2023 by the National Bureau of Statistics, a time series modelbased on ARIMA algorithmwas created to predict the urban population of China. According to the model order and parameter estimation,the parameter range is determined, and the predicted results are compared with the actual values. It is found that ARIMA(1,1,0)model is the most effective in predicting China’s urban population. Finally, comparing the ARIMA(1,1,0) model with the exponentialcurve fitting model, it is discovered that the ARIMAmodel has higher accuracy in population prediction. Themodel shows that the urbanpopulation in China will proceed to grow in the short run, but the growth trend will gradually slow down.

关 键 词：城镇人口预测 时间序列分析 ARIMA模型 曲线拟合 

分 类 号：O211.61[理学—概率论与数理统计]