灰色马尔柯夫预测模型及其应用

Grey-Markov Forecasting Model and its Application

一、引言 由于灰色预测所需信息较少,计算简便,精度较高,因此在社会经济系统的建模、分析和预测中得到广泛应用。在粮食产量和经济作物方面的应用,也取得了很好的效果。但由于灰色预测是指以GM(1,1)模型为基础所进行的预测,GM(1,1)模型的解为指数型曲线,其预测的几何图形是一条较平滑的曲线,因而对波动性较大数据列的拟合较差,预测精度较低。而马尔柯夫概率矩阵预测的研究对象是一个随机变化的动态系统,它是根据状态之间的转移概率来预测未来系统发展的。

This paper presents a grey-markov forecasting model which has the merits of both grey GM(1,1) forecast and markov transition probability matrix forecast, and would widen the applicative scope of grey forecast. In particular, the forecasted results of the model are more precise than those of other models for data sequences with heavy random fluctuation. The silkworm cocoon production in Zhejiang Province and other case studies show that the forecasting precision of the model is satisfactory.