摘要
根据我国1978~2011年的粮食产量及相关因素的历史数据,分别建立了多元回归模型、指数平滑模型、BP神经网络模型.在此基础上,根据误差平方和最小的原则,建立了基于诱导有序加权平均(IO-WA)算子的组合预测模型,经验证该模型具有较高的预测精度,并对我国未来5年的粮食产量作出了预测.
This paper established multiple regression model, exponential smoothing model and BP neural net- work model based on the historical data of grain production and related factors. And on this basis, according to the principle of minimum the sum of square errors, we built combination forecasting model based on induced ordered weighted averaging (IOWA) operators which was verified has high precision. Then, we predicted grain production in the next 5 years of our country.
出处
《菏泽学院学报》
2013年第2期14-18,共5页
Journal of Heze University
基金
国家社科基金资助项目(12BTJ008)
安徽财经大学科研项目(ACKY1315ZDB)