Advertisement

Anisotropic Matérn correlation and spatial prediction using REML

  • Kathryn A. Haskard
  • Brian R. Cullis
  • Arūnas P. Verbyla
Article

Abstract

The Matérn correlation function provides great flexibility for modeling spatially correlated random processes in two dimensions, in particular via a smoothness parameter, whose estimation allows data to determine the degree of smoothness of a spatial process. The extension to include anisotropy provides a very general and flexible class of spatial covariance functions that can be used in a model-based approach to geostatistics, in which parameter estimation is achieved via REML and prediction is within the E-BLUP framework. In this article we develop a general class of linear mixed models using an anisotropic Matérn class with an extended metric. The approach is illustrated by application to soil salinity data in a rice-growing field in Australia, and to fine-scale soil pH data. It is found that anisotropy is an important aspect of both datasets, emphasizing the value of a straightforward and accessible approach to modeling anisotropy.

Key Words

Geometric anisotropy Kriging Model-based geostatistics Residual maximum likelihood Spatial correlation 

References

  1. Abramowitz, M., and Stegun, I. A. (eds.) (1965). Handbook of Mathematical Functions, New York: Dover Publications.Google Scholar
  2. Akaike, H. (1973), “Information Theory and an Extension of the Maximum Likelihood Principle,” in Proceedings 2nd International Symposium on Information Science, Budapest, eds. B. N. Petrov and F. Csaki, Akademai Kiado, pp. 267–281.Google Scholar
  3. Beecher, H. G., Hume, I. H., and Dunn, B. W. (2002), “Improved Method for Assessing Rice Soil Suitability to Restrict Recharge,” Australian Journal or Experimental Agriculture, 42, 297–307.CrossRefGoogle Scholar
  4. Diggle, P. J., Ribeiro, P. J. J., and Christensen, O. F. (2003), “An Introduction to Model-Based Geostatistics,” in Spatial Statistics and Computational Methods, ed. J. Møller, New York: Springer, pp. 43–86Google Scholar
  5. Gilmour, A. R., Cullis, B. R., and Verbyla A. P. (1997), “Accounting for Natural and Extraneous Variation in the Analysis of Field Experiments,” Journal of Agricultural, Biological and Environmental Statistics, 2, 269–293.CrossRefMathSciNetGoogle Scholar
  6. Gilmour, A. R., Cullis, B. R., Welham, S. J., Gogel, B. J., and Thompson, R. (2004), “An Efficient Computing Strategy for Prediction in Mixed Linear Models,” Computational Statistics and Data Analysis, 44, 571–586.CrossRefMathSciNetGoogle Scholar
  7. Gilmour, A. R., Gogel, B. J., Cullis, B. R., and Thompson, R. (2005), ASReml User Guide Release 2.0, VSN International Ltd, Hemel Hempstead, HP1 1ES, UK.Google Scholar
  8. Gilmour, A. R., Thompson, R., and Cullis, B. R. (1995), “Average Information REML: An Efficient Algorithm for Variance Parameter Estimation in Linear Mixed Models,” Biometrics, 51, 1440–1450.zbMATHCrossRefGoogle Scholar
  9. Haskard, K. A. (2007), “Anisotropic Matérn Correlation and Other Issues in Model-Based Geostatistics,” unpublished Ph. D. thesis, BiometricsSA, University of Adelaide. in preparation.Google Scholar
  10. Haskard, K. A., Cullis, B. R., and Verbyla, A. P. (2007), “Estimation and Sampling Issues for an Anisotropic Matérn Correlation Model,” submitted.Google Scholar
  11. Haskard, K. A., and Verbyla, A. P. (2007), “A Note on an Apparent Non-regular Hypothesis Test,” submitted.Google Scholar
  12. Kenward, M. G., and Roger, J. H. (1997), “Small Sample Inference for Fixed Effects From Restricted Maximum Likelihood,” Biometrics, 53, 983–997.zbMATHCrossRefGoogle Scholar
  13. Kitanidis, P. K. (1983), “Statistical Estimation of Polynomial Generalized Covariance Functions and Hydrological Applications,” Water Resources Research, 19, 909–921.CrossRefGoogle Scholar
  14. Lark, R. M., and Cullis, B. R. (2004), “Model-Based Analysis Using REML for Inference from Systematically Sampled Data on Soil,” European Journal of Soil Science, 55, 799–813.CrossRefGoogle Scholar
  15. Matheron, G. (1973), “The Intrinsic Random Functions and their Application,” Journal of Applied Probability, 5, 439–468.zbMATHCrossRefMathSciNetGoogle Scholar
  16. Stein, M. L. (1999), Interpolation of Spatial Data: Some Theory for Kriging, New York: Springer-Verlag.zbMATHGoogle Scholar
  17. Stein, M. L., Chi, Z., and Welty, L. J. (2004), “Approximating Likelihoods for Large Spatial Datasts,” Journal of the Royal Statistical Society, Series B, 66, 275–296.zbMATHCrossRefMathSciNetGoogle Scholar
  18. Stram, D. O., and Lee, J. W. (1994), “Variance Components Testing in the Longitudinal Mixed Effects Model,” Biometrics, 50, 1171–1177.zbMATHCrossRefGoogle Scholar
  19. Stram, D. O., and Lee, J. W. (1994), “Variance Components Testing in the Longitudinal Mixed Effects Model,” Biometrics, 50, 1171–1177.zbMATHCrossRefGoogle Scholar
  20. Vecchia, A. V. (1998), “Estimation and Model Identification for Continuous Spatial Processes,” Journal of the Royal Statistical Society, Series B, 50, 297–312.MathSciNetGoogle Scholar
  21. Verbyla, A. P. (1990), “A Conditional Derivation of Residual Maximum Likelihood,” Australian Journal of Statistics, 32, 227–230.CrossRefGoogle Scholar
  22. Webster, R., and Oliver, M. A. (2001), Geostatistics for Environmental Scienctists, Chichester: Wiley.Google Scholar

Copyright information

© International Biometric Society 2007

Authors and Affiliations

  • Kathryn A. Haskard
    • 1
  • Brian R. Cullis
    • 2
  • Arūnas P. Verbyla
    • 3
  1. 1.BiometricsSASARDIAdelaideAustralia
  2. 2.New South Wales Department of Primary IndustriesWagga WaggaAustralia
  3. 3.School of Agriculture, Food and WineThe University of AdelaideGlen OsmondAustralia

Personalised recommendations