针对单项预测存在一定随机性、预测精度较低等问题,基于误差平方和最小准则,结合 GIOWA算子提出 4 种特殊参数的变权系数组合模型,以 2000—2020 年安徽省城镇居民人均可支配收入数据为样本,对安徽省未来 5 年城镇居民人均可支配收入进行预测。 首先分别应用 3 种单项预测模型对安徽省城镇居民人均可支配收入进行拟合预测,然后以误差平方和最小为准则,结合 GIOWA 算子构建变权系数组合模型,同时对 GIOWA 算子取 4 种特殊参数得到相应的组合预测模型,最后应用所构建模型对安徽省未来 5 年城镇居民人均可支配收入进行预测。 结果表明:变系数组合预测模型预测效果优于单项预测模型;安徽省未来 5 年城镇居民人均可支配收入将会持续稳定增长。
Aiming at the problems of certain randomness and low prediction accuracy of single prediction, based on the minimum sum of square error criterion, combined with the GIOWA operator, a combination model of four special parameters with variable weight coefficients is proposed. Taking the data of the per capita disposable income of urban residents in Anhui Province from 2000 to 2020 as a sample, a combined forecasting model based on the sum of squares of errors and the GIOWA operator is constructed to predict the per capita disposable income of urban residents in Anhui Province in the next 5 years. Firstly, three single forecast models are applied to fit and forecast the per capita disposable income of urban residents in Anhui Province. Then the study uses the minimum sum of square error as the criterion, combined with the GIOWA operator to construct a variable weight coefficient combination model, and at the same time takes four special parameters for the GIOWA operator to obtain the corresponding combination prediction model. Finally, the GIOWA combined prediction model is used to predict the per capita disposable income of urban residents in Anhui Province in the next 5 years. The results show that the prediction effect of the variable weight coefficient combination forecasting model is better than that of the single prediction model; the per capita disposable income of urban residents in Anhui Province will continue to grow steadily in the next 5 years.
张文扬, 汪 凯, 袁宏俊.安徽省城镇居民人均可支配收入组合预测[J].重庆工商大学学报（自然科学版）,2022,39(5):93-104
ZHANG Wen-yang, WANG Kai, YUAN Hong-jun. Combination Forecast of Per Capita Disposable Income of Urban Residents in Anhui Province[J]. Journal of Chongqing Technology and Business University(Natural Science Edition）,2022,39(5):93-104