Parameter Estimation of Variogram Model by Weighted Total
Least Squares Regression
1 Faculty of Geomatics, East China Institute of Technology, 418 Guanglan Road, Nanchang 330013, China
2 Key Laboratory of Watershed Ecology and Geographical Environment Monitoring, NASMG,
418 Guanglan Road, Nanchang 330013, China
3 Jiangxi Province Key Lab for Digital Land, 418 Guanglan Road, Nanchang 330013, China
Abstract:Considering errors of distances, weighted total least squares regression is applied to estimate parameters of variogram model. According to variance-covariance propagation law, variogram values and distances after classification are found unequally accurate. Combined with two weighting methods of variogram values, the entropy weight method and number of points method, the weight method of distances derived by the variance-covariance propagation law is used to estimate parameters. Taking the power function model as an example, weighted total least squares regression is proven to be more reasonable and accurate by results of simulated and actual data.