基于预测有效度成分数据处理方法组合预测  

Component Data Processing Method Combination Forecast Based on Predictive Validity

作　　者：房玄骅 王艺宁 刘夕 FANG Xuan-hua;WANG Yi-ning;LIU Xi(Economics School of Anhui University,Hefei 230601,China;School of Mathematical Sciences,Anhui University,Hefei 230601,China)

机构地区：[1]安徽大学经济学院,合肥230601 [2]安徽大学数学科学学院,合肥230601

出　　处：《价值工程》2019年第21期192-194,共3页Value Engineering

基　　金：安徽大学本科生创新创业研究项目(No.201810357203,201810357518,201810357519)

摘　　要：为提高成分数据时序预测准确性,提出一种以二阶预测有效性作标准的多种数据处理方法的组合预测。选择成分数据的多种数据转化方法,将有约束时序用对数比,中心对数,超球面变换方法转换成无约束时序后,利用ARIMA-ANN模型对转换后无约束时序预测,对结果做反变换,恢复为成分数据得单项预测结果。最后对得到的单项预测结果进行基于二阶预测有效度的加权几何平均组合,得到相对最优的组合预测结果。In order to improve the accuracy of time series prediction of component data, a combined prediction of multiple data processing methods based on second-order prediction validity is proposed. Selecting a variety of data transformation methods for component data,after the constrained time series is transformed into the unconstrained time series by the logarithmic ratio, the central logarithm and the hypersphere transformation method, the ARIMA-ANN model is used to predict the unconstrained time series after the transformation, and the result is inversely transformed to restore the component data to a single prediction. result Finally, the weighted geometric mean combination based on the second-order prediction validity is obtained for the obtained single prediction result, and the relatively optimal combined prediction result is obtained.

关 键 词：成分数据 组合预测 预测有效度 对数比变换 中心对数变换 超球面变换 

分 类 号：O221.1[理学—运筹学与控制论]