灰色多元变权组合预测模型及其应用  

Grey Multivariate Variable Weight Combination Prediction Model and Its Application

作　　者：吕欣曼 殷克东 李雪梅[1,2] Lyu Xinman;Yin Kedong;Li Xuemei(School of Economics,Ocean University of China,Qingdao Shandong 266100,China;Institute of Marine Development,Ocean University of China,Qingdao Shandong 266100,China;Institute of Marine Economics and Management,Shandong University of Finance and Economics,Jinan 250014,China)

机构地区：[1]中国海洋大学经济学院,山东青岛266100 [2]中国海洋大学海洋发展研究院,山东青岛266100 [3]山东财经大学海洋经济与管理研究院,济南250014

出　　处：《统计与决策》2022年第14期25-29,共5页Statistics & Decision

基　　金：国家自然科学基金资助项目(41701593);山东省社会科学规划研究项目(16DJJJ06);青岛市社会科学规划研究项目(QDSKL1601014)。

摘　　要：为了进一步提高预测精度,考虑多因素对经济社会的影响,文章基于OGM(1,N)、BP神经网络和偏最小二乘回归的传统预测模型,利用方差倒数法构建了一种变权多元组合预测模型,并针对中国二氧化碳排放量进行预测。实证结果显示,灰色多元组合预测模型的预测精度高达99.764%,能有效地弱化三种单项模型的缺点,具有很好的预测性能。根据预测结果,预计到2025年中国低碳发展取得成效,碳排放量增速放缓,以1%的速度继续保持增长。In order to further improve the prediction accuracy,this paper considers the influence of multiple factors on the economy and society.Then,based on traditional prediction models of OGM(1,N),BP neural network and partial least squares re-gression,the paper uses variance reciprocal method to construct a variable weight multivariate combination prediction model,and finally predicts carbon dioxide emissions in China.The empirical results show that the prediction accuracy of the grey multiple combination prediction model is as high as 99.764%,which can effectively weaken the shortcomings of the three single models and has a good prediction performance.According to the prediction results,China’s low-carbon development is expected to achieve results by 2025,with the growth rate of carbon emissions slowing down and continuing to grow at a rate of 1%.

关 键 词：灰色多元变权预测 BP神经网络 偏最小二乘回归分析 组合预测 

分 类 号：F205[经济管理—国民经济]