GM(1,1)模型对烟草产量的灰色预测

The forecast of the tobacco output by using GM(1,1) model

为了科学规划福建省宁化县烟草种植，运用灰色系统理论对该县1996～2004年烟草产量数据进行分析，建立了灰色系统的预测GM（1，1）模型。运用GM（1，1）模型对全县的烟草产量进行预测，并对预测结果分别进行了残差、关联度和后验差检验。结果表明，预测产量与实际产量大体吻合，二者的平均相对误差为2．28％，关联度为0．91（k=0．5），方差比为0．4，小误差概率为1。经与检验标准对比，显示所建模型精度较高，可用于宁化县烟草产量的短期预测。

Based on the tobacco output data of Ninghua county, Fujian province, GM（ 1,1 ）model is established to forecast the tobacco output of the county. With the forecast results of the model,the residual, degree of collaboration and error inspection of post-sample are also tested. The results show that the average relative error of established model is 2.28% , the degree of collaboration of that is 0.91 （k=0.5）and the minimal error of that is 1. It imply that the established model has high precision and can be used as a tool to forecast the tobacco output of Ninghua county.