摘要
为了分析新疆城镇居民人均可支配收入的变化趋势,利用1995~2018年新疆城镇居民人均可支配收入的数据,分别建立了ARIMA模型和指数平滑模型.利用这两个单项模型建立了城镇居民人均可支配收入的组合模型.实例结果表明:指数平滑模型的拟合效果优于ARIMA模型,而组合模型的拟合效果优于两个单项模型.最后,运用组合模型对未来五年新疆城镇居民人均可支配收入进行预测.
In order to analyze the change trend of per capita disposable income of urban residents in Xinjiang,ARIMA model and exponential smoothing model are established respectively by using the data of per capita disposable income of urban residents in Xinjiang from 1995 to 2018.And a combined model of per capita disposable income of urban residents is established based on the two single models.It is showed in the results that the fitting effect of exponential smoothing model is better than that of ARIMA model,and that of combination model is better than that of two single models.Finally,the per capita disposable income of urban residents in Xinjiang in the next five years is predicted based on the combined model.
作者
夏春芳
李树
王丰效
XIA Chun-fang;LI Shu;WANG Feng-xiao(School of Mathematics and Statistics,Kashi University,Kashi 844000,Xinjiang,China)
出处
《兰州文理学院学报(自然科学版)》
2020年第3期24-27,共4页
Journal of Lanzhou University of Arts and Science(Natural Sciences)
基金
国家自然科学基金(10771003)。
关键词
时间序列
ARIMA
指数平滑
组合预测
time series
ARIMA
exponential smoothing
combination forecasting