用灰色理论预测青海湖裸鲤的年产量

Application of the grey prediction model to forecast Gymnocypris przewalskii przewalskii production

应用灰色理论与方法，以1991—1998年青海湖裸鲤Gymnocypris przewalskii przewalskii的年产量统计数据为基础，建立了灰色系统理论GM（1，1）预测模型，用该模型对1999年青海湖裸鲤的年产量进行了预测。结果表明：青海湖裸鲤年产量的时间响应函数模型为χ^（0）（k＋1）=4974．9670996e^-0.232119k，多年平均相对误差为10．33％，后验差比值C=0．248352，小误差频率P=1，模型的预测精度达到一级；1999年青海湖裸鲤年产量的预测值为776．8t，与实际产量（807t）的相对误差为3．73％，模型的预测效果比较理想。

The theory and methods of Grey System were used to establish the Forecasting Model of GM （ 1,1 ） to predict C, ymnocypris przewalskii przewalskii production on the basis of fishery data of C, ymnocypris przewalski przewalskiifrom 1991 to 1998. The GM （ 1,1 ） prediction fishery yield model was established asχ^（0）（k＋1）=4974.9670996e^-0.232119k The mean error of regression testing was 10.33%. The difference check of the prediction model： C =0. 248352,P = 1 and the model precision ranks the first class. The predicted yield of the model in 1999 was about 776.8 t, with a relative error of 3.73% to actual yield. The model was proved to be feasible and suitable in this paper.