由于易受气候以及自然灾害等因素的影响,小麦产量的变化往往呈现出接近周期性的非线性趋势,为了克服传统GM(1,1)模型预测结果呈单调性的缺点,本文先将原始数据 x (0)映射成正弦函数 x (0)=siny (0)( y (0)=arcsinx (0))后的y (0)值作为新的原始数据带入GM(1,1)模型,然后再将GM(1,1)模型的预测值运用x (0)=siny (0)还原回去,对GM(1,1)预测模型加以改进.结果表明,在预测准确度上,正弦函数变换型的GM(1,1)模型明显优于传统的GM(1,1)模型.%Due to the influence factors of climate and other natural disasters,the changes of wheat yield often pres-ents a nonlinear trend close to periodically,in order to overcome the short comigs that the results predicted by the traditional GM(1,1) model show the monotonicity,this paper first will map the original data x (0)into sine function x (0) =siny (0)( y (0)=arcsinx (0)) the value as the new raw data into GM(1,1) model,then the GM(1,1) model pre-dicted by x (0)=siny (0)reduction back,GM(1,1) prediction model is improved.The experimental results show that the GM(1,1) model of sinusoidal function transformation is superior to the traditional GM(1,1) model in prediction accuracy.
展开▼