运用改进的GM(1,1)模型预测我国猪饲料产量

Prediction of Pig Feed Production of China Based on Improved GM(1,1)Model

针对我国猪饲料产量数据序列分布呈非单调性的特点,不符合传统灰色GM(1,1)模型建模条件,预测效果不佳,对GM(1,1)模型进行了改进。首先,通过波形数据生成法(BX)将不规则的数据序列转化为单调数据序列,然后用新生成的数据序列建模进行预测。结果显示,改进GM(1,1)模型比传统GM(1,1)模型的平均预测误差减小了29.4689%,由改进模型得到2021年我国猪饲料产量为9305.296万吨,这一预测值也比传统GM(1,1)预测值更可靠。

In view of the non monotonicity of the data series distribution of pig feed yield in China,the GM(1,1)model is not suitable for the traditional GM(1,1)model,and the prediction effect is not good.The GM(1,1)model is improved.Firstly,the irregular data sequence is transformed into smooth and monotonic data sequence by the method of waveform data generation(BX),and then the new data sequence is used to model and predict.The results showed that the average prediction error of the improved GM(1,1)model was 29.4689%less than that of the traditional GM(1,1)model,and the pig feed output in 2021 was 93.05296 million tons,which was more reliable than that of the traditional GM(1,1)model.