Mathematical Geology

, Volume 39, Issue 2, pp 247–260 | Cite as

Hybrid Estimation of Semivariogram Parameters

  • Hao ZhangEmail author
  • Dale L. Zimmerman

Two widely used methods of semivariogram estimation are weighted least squares estimation and maximum likelihood estimation. The former have certain computational advantages, whereas the latter are more statistically efficient. We introduce and study a “hybrid” semivariogram estimation procedure that combines weighted least squares estimation of the range parameter with maximum likelihood estimation of the sill (and nugget) assuming known range, in such a way that the sill-to-range ratio in an exponential semivariogram is estimated consistently under an infill asymptotic regime. We show empirically that such a procedure is nearly as efficient computationally, and more efficient statistically for some parameters, than weighted least squares estimation of all of the semivariogram’s parameters. Furthermore, we demonstrate that standard plug-in (or empirical) spatial predictors and prediction error variances, obtained by replacing the unknown semivariogram parameters with estimates in expressions for the ordinary kriging predictor and kriging variance, respectively, perform better when hybrid estimates are plugged in than when weighted least squares estimates are plugged in. In view of these results and the simplicity of computing the hybrid estimates from weighted least squares estimates, we suggest that software that currently estimates the semivariogram by weighted least squares methods be amended to include hybrid estimation as an option.


consistency geostatistics kriging maximum likelihood estimation weighted least squares estimation 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Clark, I., 1979, Practical geostatistics: Applied Science Publishers: Essex, England, 129 p.Google Scholar
  2. Cressie, N., 1985, Fitting variogram models by weighted least squares: Math. Geol., v. 17, no. 5, p. 563–586.CrossRefGoogle Scholar
  3. Curriero, F. C., and Lele, S., 1999, A composite likelihood approach to semivariogram estimation: Journ. of Agricultural, Biological, and Environmental Stat., v. 4, no. 1, p. 9–28.CrossRefGoogle Scholar
  4. Dubin, R. A., 1994, Estimating correlograms: A Monte Carlo study: Geographical Systems, v. 1, p. 189–202.Google Scholar
  5. Ecker, M. D., and Gelfand, A. E., 1997, Bayesian variogram modeling for an isotropic spatial process: Journal of Agricultural, Biological, and Environmental Stat., v. 2, no. 4, p. 347–369.CrossRefGoogle Scholar
  6. Ecker, M. D., and Gelfand, A. E., 1999, Bayesian modeling and inference for geometrically anisotropic spatial data: Math. Geol., v. 31, no. 1, p. 67–83.Google Scholar
  7. Genton, M. G., 1998, Variogram fitting by generalized least squares using an explicit formula for the covariance structure: Math. Geol., v. 30, no. 4, p. 323–345.CrossRefGoogle Scholar
  8. Handcock, M. S., and Stein, M. L., 1993, A Bayesian analysis of kriging: Technometrics, v. 35, no. 4, p. 403–410.CrossRefGoogle Scholar
  9. Journel, A. G., and Huijbregts, C. J., 1978, Mining geostatistics: Academic Press, London, 600 p.Google Scholar
  10. Kitanidis, P. K., 1983, Statistical estimation of polynomial generalized covariance functions and hydrologic applications: Water Resources Res., v. 19, no. 4, p. 909–921.CrossRefGoogle Scholar
  11. Mardia, K. V., and Marshall, R. J., 1984, Maximum likelihood estimation of models for residual covariance in spatial regression: Biometrika, v. 71, no. 1, p. 135–146.CrossRefGoogle Scholar
  12. Stein, M. L., 1988, Asymptotically efficient spatial interpolation with a misspecified covariance function: Annals of Stat., v. 16, no. 1, p. 55–63.Google Scholar
  13. Stein, M. L., 1990, Uniform asymptotic optimality of linear predictions of a random field using an incorrect second-order structure: Annals of Stat., v. 18, no. 2, p. 850–872.Google Scholar
  14. Stein, M. L., and Handcock, M. S., 1989, Some aymptotic properties of kriging when the covariance function is misspecified: Math. Geol., v. 21, no. 2, p. 171–190.CrossRefGoogle Scholar
  15. Stein, M. L., 1999, Interpolation of spatial data: Springer, New York, 247 p.Google Scholar
  16. Ying, Z., 1991, Asymptotic properties of a maximum likelihood estimator with data from a Gaussian process: Journal of Multivariate Analysis, v. 36, no. 2, p. 280–296.CrossRefGoogle Scholar
  17. Zhang, H., 2004, Inconsistent estimation and asymptotically equal interpolations in model-based geostatistics: Jour. of the Am. Stat. Association, v. 99, no. 465, p. 250–261.CrossRefGoogle Scholar
  18. Zimmerman, D. L., 1989, Computationally efficient restricted maximum likelihood estimation of generalized covariance functions: Math. Geol., v. 21, no. 7, p. 655–672.CrossRefGoogle Scholar
  19. Zimmerman, D. L., and Zimmerman, M. B., 1991, A comparison of spatial semivariogram estimators and corresponding ordinary kriging predictors: Technometrics, v. 33, no. 1, p. 77–91.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2007

Authors and Affiliations

  1. 1.Department of StatisticsWashington State UniversityPullmanUSA
  2. 2.Department of Statistics and Actuarial ScienceUniversity of IowaIowa CityUSA

Personalised recommendations