Semivariogram modeling by weighted least squares
Semivariogram modeling by weighted least squares
复制标题
DOI：
10.1016/0098-3004(95)00095-x
复制
发表时间：
1996-05-01
影响因子：
4.4
通讯作者：
Yu, YS
中科院分区：
文献类型：
--
作者：
Jian, XD;Olea, RA;Yu, YS
允许的半变异函数模型是具有连续时空变化的属性的地质统计估计和模拟的基础。通常的做法是手动将这些模型拟合到实验半方差图中。加权最小二乘拟合可以在更短的时间内系统地产生与人工拟合相当的结果，并为替代模型的适当比较提供了赤池信息准则。通过简单嵌套模型的拟合实例，说明了计算机程序的应用。版权所有爱思唯尔科学有限公司
Permissible semivariogram models are fundamental for geostatistical estimation and simulation of attributes having a continuous spatiotemporal variation. The usual practice is to fit those models manually to experimental semivariograms. Fitting by weighted least squares produces comparable results to fitting manually in less time, systematically, and provides an Akaike information criterion for the proper comparison of alternative models. We illustrate the application of a computer program with examples showing the fitting of simple and nested models. Copyright (C) 1996 Elsevier Science Ltd