基于DGM的模糊多目标区间预测

Fuzzy Multi-objective Interval Forecasting Based on a Discrete Grey Model

无论对预测过程还是决策过程,区间预测相对于点估计预测往往拥有更重要的信息量。本文基于模糊多目标规划模型和离散灰色模型理论(DGM),给出了基于DGM的模糊多目标区间预测模型。同时,为了有利于比较区间预测的优劣,又提出了评价区间预测优劣的两个指标:区间中心的均值相对百分比误差和区间边界的均值相对百分比误差。最后,将LCD TV销售量作为本模型的实验数据,给出了区间预测的结果,并将我们的结果与其他两个模型的预测结果进行了比较分析。

Interval forecasting gives more important information during whatever forecasting processes and decision-making processes. An improved interval forecasting model based on fuzzy multi-objective programming combined with discrete grey model theory （abbreviated DGM） is proposed. At the same time, Mean Relative Percentage Error of Interval Center （MRPEC） and Mean Relative Percentage Error of Interval Boundary （MRPEB） as two evaluation indexes are given in order to compare the accuracy of the estimate interval forecasting. Finally, the LCD TV sales is used as the experimental data for the forecasting model. The results of the interval forecasting are given to show that the proposed model is an efficient interval forecasting model compared with the other two forecasting models.