基于ARIMA-GM(1,1)组合模型对居民人均可支配收入的预测分析

Forecast and analysis of per capita disposable income based on ARIMA-GM(1,1)combination model

以1983—2022年中国居民人均可支配收入作为研究数据,通过ARIMA模型、灰度预测GM(1,1)模型及其相关组合预测模型对该时间序列数据进行预测分析.以2018—2022年数据作为模型测试集,引入相关评价指标进行拟合优度分析,给出了2023年居民人均可支配收入的最佳预测结果.结果表明,所有预测模型的MAPE均低于1%,并且使用算术平均法的ARIMA-GM(1,1)组合预测模型的预测精度明显高于单一预测模型,拟合效果最佳.

The per capita disposable income of Chinese residents from 1983 to 2022 is taken as the research data,and the time series data are predicted and analyzed by ARIMA model,grayscale prediction GM(1,1)model and its related combination prediction model.The data from 2018 to 2022 is taken as the model test set,relevant evaluation indexes are introduced for goodness of fit analysis,and the best prediction result of per capita disposable income in 2023 is given.The results show that the MAPE of all prediction models is lower than 1%,and the prediction accuracy of the combined prediction model based on ARIMA-GM(1,1)using arithmetic average method is significantly higher than that of the single prediction model,and the fitting effect is the best.