一类离散灰色模型及其预测效果研究

Research on discrete grey model and its predictive result

GM(1,1)模型是灰色系统理论中最重要的内容之一,针对该模型在预测时出现的预测精度问题进行讨论,进而建立了新的离散灰色模型:始点固定离散灰色模型(SDGM)和终点固定离散灰色模型(EDGM),运用大量的数据模拟和预测,将新建立的2个模型和GM(1,1)模型的进行效果比较,发现新建立的2个模型均比GM(1,1)模型有更高的模拟精度和预测精度.

The forecasting model of GM（1,1） is one of the most important contents of the grey system theory. In this paper, we discuss the problem of forecasting precision which is often appeared in the model. Then we create two new discrete grey models： starting-point fixed discrete grey model（SDGM） and ending-point fixed discrete grey model（EDGM）. We did simulation and predication by utilizing a large amount of data. By contrasting the results using the two new models with that using GM（1,1） model, it is indicated that we can get more accurate results in the two new models.