泛Kriging和广义线性回归模型是空间对象属性值预测最常用的方法,两种模型各有侧重。本文结合这两种模型的优点,并针对实际问题将两种模型合并,形成新的空间广义线性回归模型,并通过混合模型进行参数估计。实验表明,采用空间广义线性回归预测模型能获得更高的预测精度。 Pan Kriging and generalized linear regression models are the most commonly used methods for predicting the attribute values ​​of spatial objects, and each model has its own focus. Combining the advantages of these two models, this paper combines the two models for practical problems to form a new generalized linear regression model of space, and estimates the parameters through the hybrid model. Experiments show that using generalized linear regression prediction model can obtain higher prediction accuracy.