机构地区: 中国农业科学院中国水稻研究所
出 处: 《生物数学学报》 1997年第S1期564-568,共5页
摘 要: 文章首先讨论了模型的拟合效果和预测效果的区别,给出了相应的估算方法,同时也讨论了抽样观测值大小对预测精度的影响,并结合实例说明如何从一组模型中选择一预测效果最好的模型.对1992年中国水稻研究所富阳实验基地和绍兴县测产样本点的数据,得到预测效果较好的模型为:产量预测值=理论产量+线性纠正估计,其中线性纠正的形式对两组数据略有不同. The difference between the fitted models and the prediction models are presented,the corresponding accuracy estimation methods are provided.Also the effect of the observations on the prediction accuracy is discussed.One example is given to show how to choose the best prediction model from available models.The best prediction model for the data in 1992 from Fuyang county and Shaoxing County of Zhejiang Province share the same model form: yield prediction=theoretical yield+linear adjustment.