Fitting variogram models by weighted least squares

  • Noel Cressie
Article

Abstract

The method of weighted least squares is shown to be an appropriate way of fitting variogram models. The weighting scheme automatically gives most weight to early lags and down-weights those lags with a small number of pairs. Although weights are derived assuming the data are Gaussian (normal), they are shown to be still appropriate in the setting where data are a (smooth) transform of the Gaussian case. The method of (iterated) generalized least squares, which takes into account correlation between variogram estimators at different lags, offer more statistical efficiency at the price of more complexity. Weighted least squares for the robust estimator, based on square root differences, is less of a compromise.

Key words

generalized least squares kriging median polish robustness stationarity 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Armstrong, M., 1984, Improving the estimation and modelling of the variogram,in G. Verly et al. (Eds.) Geostatistics for natural resources characterization: Dordrecht, Reidel, p. 1–19.Google Scholar
  2. Armstrong, M. and Delfiner, P., 1980, Towards a more robust variogram: A case study on coal, Internal note N-671: Centre de Geostatistique, Fontainebleau, France.Google Scholar
  3. Buxton, B., 1982, A geostatistical case study, unpublished masters thesis: Stanford University, California, 84 p.Google Scholar
  4. Carroll, R. J. and Ruppert, D., 1982, A comparison between maximum likelihood and generalized least squares in a heteroscedastic linear model:Jour. Amer. Stat. Assoc., v. 77, p. 878–882.Google Scholar
  5. Clark, I., 1979, Practical geostatistics: Applied Science Publishers, Essex, England, 129 p.Google Scholar
  6. Cressie, N. A. C., 1979, Straight line fitting and variogram estimation (with discussion):Bull. Inter. Stat. Inst., v. 48, Book 3, p. 573–582.Google Scholar
  7. Cressie, N. A. C., 1980, M-estimation in the presence of unequal scale.Stat. Neerlandica, v. 34, p. 19–32.Google Scholar
  8. Cressie, N. A. C., 1984, Towards resistant geostatistics, in G. Verley et al. (Eds.) Geostatistics for natural resources characterization: Dordrecht, Reidel, p. 21–44.Google Scholar
  9. Cressie, N. A. C. and Glonek, G., 1984, Median based covariogram estimators reduce bias:Stat. Prob. Lett., v. 2, p. 299–304.Google Scholar
  10. Cressie, N. A. C. and Hawkins, D. M., 1980, Robust estimation of the variogram:Jour. Inter. Assoc. Math. Geol., v. 12, p. 115–125.Google Scholar
  11. David, M., 1977, Geostatistical ore reserve estimation: Amsterdam, Elsevier, 364 p.Google Scholar
  12. Davis, B. M. and Borgman, L. E., 1982, A note on the asymptotic distribution of the sample variogram:Jour. Inter. Assoc. Math. Geol., v. 14, p. 189–193.Google Scholar
  13. Delfiner, P., 1976, Linear estimation of nonstationary spatial phenomena, in M. Guarascio et al. (Eds.) Advanced geostatics in the mining industry: Dordrecht, Reidel, p. 49–68.Google Scholar
  14. Diamond, P. and Armstrong, M., 1983, Robustness of variograms and conditioning of kriging matrices: Internal note N-804, Centre de Geostatistique, Fontainebleau, France.Google Scholar
  15. Gomez, M. and Hazen, K., 1970, Evaluating sulfur and ash distribution in coal seams by statistical response surface regression analysis: U.S. Bureau of Mines Report, RI 7377.Google Scholar
  16. Hawkins, D. M. and Cressie, N. A. C., 1984, Robust kriging—a proposal: Jour. Inter. Assoc. Math. Geol., v. 16, p. 3–18.Google Scholar
  17. Huber, P. J., 1964, Robust estimation of a location parameter: Ann. Math. Stat., v. 35, p. 73–101.Google Scholar
  18. Journel, A. and Huijbregts, C., 1978, Mining geostatistics: London, Academic Press, 600 p.Google Scholar
  19. King, H. F., McMahon, D. W., and Bujtor, G. J., 1982, A guide to the understanding of ore reserve estimation,in Proceedings of the Australasian Institute of Mining and Metallurgy, Supplement: v. 281, p. 1–21.Google Scholar
  20. Matheron, G., 1963, Principles of geostatistics: Econ. Geol., v. 58, p. 1246–1266.Google Scholar
  21. Matheron, G., 1971, The theory of regionalized variables and its applications: Cahiers du Centre de Morphologie Mathematique, No. 5, Fontainebleau, France.Google Scholar
  22. Sharp, W. E., 1982, Estimation of semi-variograms by the maximum entropy method:Jour. Inter. Assoc. Math. Geol., v. 14, p. 456–474.Google Scholar
  23. Starks, T. and Fang, J., 1982a, On the estimation of the generalized covariance function: Jour. Inter. Assoc. Math. Geol., v. 14, p. 57–64.Google Scholar
  24. Starks, T. and Fang, J., 1982b, The effect of drift on the experimental semi-variogram: Jour. Inter. Assoc. Math. Geol., v. 14, p. 309–320.Google Scholar
  25. Switzer, P., 1984, Inference for spatial autocorrelation functions in G. Verly et al. (Eds.) Geostatistics for natural resources characterization: Dordrecht, Reidel, p. 127–140.Google Scholar

Copyright information

© Plenum Publishing Corporation 1985

Authors and Affiliations

  • Noel Cressie
    • 1
  1. 1.Department of StatisticsIowa State UniversityAmesUSA

Personalised recommendations