Variogram Fitting Based on the Wilcoxon Norm

Within geostatistics research, effective estimation of
the variogram points has been examined, particularly in developing
robust alternatives. The parametric fit of these variogram points which
eventually defines the kriging weights, however, has not received
the same attention from a robust perspective. This paper proposes
the use of the non-linear Wilcoxon norm over weighted non-linear
least squares as a robust variogram fitting alternative. First, we
introduce the concept of variogram estimation and fitting. Then, as
an alternative to non-linear weighted least squares, we discuss the
non-linear Wilcoxon estimator. Next, the robustness properties of the
non-linear Wilcoxon are demonstrated using a contaminated spatial
data set. Finally, under simulated conditions, increasing levels of
contaminated spatial processes have their variograms points estimated
and fit. In the fitting of these variogram points, both non-linear
Weighted Least Squares and non-linear Wilcoxon fits are examined
for efficiency. At all levels of contamination (including 0%), using
a robust estimation and robust fitting procedure, the non-weighted
Wilcoxon outperforms weighted Least Squares.

Authors:

• Hazem Al-Mofleh
• John Daniels
• Joseph McKean

Keywords:

• Non-Linear Wilcoxon
• robust estimation
• Variogram estimation.

References:

[1] F. R. Hampel, “Robust estimation: A condensed partial survey,”
Zeitschrift fr Wahrscheinlichkeitstheorie und Verwandte Gebiete, vol. 27,
no. 2, pp. 87–104, 1998.
[2] P. J. Huber, Robust Statistical Procedures, 2nd ed. Philadelphia: Society
for Industrial and Applied Mathematics, 1996.
[3] G. Matheron, Trait de gostatistique applique, ser. M´emoires du Bureau
de Recherches G´eologiques et Mini`eres. ´ Editions Technip, 1963, no.
v. 2.
[4] N. Cressie and D. M. Hawkins, “Robust estimation of the variogram:
I,” Mathematical Geology, vol. 12, no. 2, pp. 115–125, 1980.
[5] N. Cressie, Statistics for Spatial Data. New York: John Wiley and
Sons, Ltd, 1991.
[6] M. G. Genton, “Highly robust variogram estimation,” Mathematical
Geology, vol. 30, no. 2, pp. 213–221, 1998.
[7] K. V. Mardia and R. J. Marshall, “Maximum likelihood estimation of
models for residual covariance in spatial regression,” Biometrika, vol. 71,
no. 1, pp. 135–146, 1984.
[8] H. Patterson and R. Thompson, “Recovery of inter-block information
when block sizes are unequal,” Biometrika, vol. 58, no. 3, pp. 545–554,
1971.
[9] “Maximum likelihood estimation of components of variance.”
Constanta, Romania: 8th International Biometrics Conference, 1974,
pp. 197–207.
[10] C. R. Rao, “Minqe theory and its relation to ml and mml estimation of
variance components,” Sankhy: The Indian Journal of Statistics, Series
B (1960-2002), vol. 41, no. 3/4, pp. 138–153, 1979.
[11] N. Cressie, “Fitting variogram models by weighted least squares,”
Journal of the International Association for Mathematical Geology,
vol. 17, no. 5, pp. 563–586, 1985.
[12] T. P. Hettmansperger and J. W. McKean, Robust Nonparametric
Statistical Methods, 2nd ed. New York: Chapman-Hall, 2011.
[13] J. W. McKean and J. D. Kloke, “Efficient and adaptive rank-based
fits for linear models with skew-normal errors,” Journal of Statistical
Distributions and Applications, vol. 1, no. 1, pp. 1–18, 2014.
[14] A. Abebe and J. W. McKean, “Highly efficient nonlinear regression
based on the wilcoxon norm,” in D. Umbach (ed.), Festschrift in Honor
of Mir Masoom Ali on the Occasion of his Retirement, Ball State
University, Muncie, IN, 2007, pp. 340–357.
[15] H. L. Koul, G. L. Sievers, and J. W. McKean, “An estimator of the
scale parameter for the rank analysis of linear models under general
score functions,” Scandinavian Journal of Statistics, vol. 14, no. 2, pp.
131–141, 1987.
[16] J. D. Kloke and J. W. McKean, “Rfit: Rank-based estimation for linear
models,” The R Journal, vol. 4, no. 2, pp. 57–64, 2012.
[17] R. G. Clark and S. Allingham, “Robust resampling confidence intervals
for empirical variograms,” Mathematical Geosciences, vol. 43, no. 2,
pp. 243–259, 2011.
[18] R Core Team, R: A language and environment for statistical computing,
R Foundation for Statistical Computing, Vienna, Austria, 2015, r
package version 3.2.2. (Online). Available: https://www.R-project.org/
[19] L. Tanga, W. R. Schucany, W. A. Woodwardb, and R. F. Gunstb, “A
parametric spatial bootstrap,” Southern Methodist University, Dallas,
Texas, Tech. Rep. SMU-TR-337, 2006.
[20] H. M. Al-Mofleh, J. E. Daniels, and J. W. McKean, “Robust variogram
fitting using non-linear rank-based estimators,” International Journal
of Mathematical, Computational, Physical, Electrical and Computer
Engineering, vol. 10, no. 2, pp. 68 – 77, 2016. (Online). Available:
http://waset.org/Publications?p=110
[21] J.-P. Chiles and P. Delfiner, Geostatistics: Modelling Spatial Uncertainty.
New York: John Wiley and Sons, Ltd, 1999.
[22] S. Garrigues, D. Allard, F. Baret, and M. Weiss, “Quantifying spatial
heterogeneity at the landscape scale using variogram models,” Remote
Sensing of Environment, vol. 103, no. 1, pp. 81–96, 2006.
[23] O. Atteia, J.-P. Dubois, and R. Webster, “Geostatistical analysis of soil
contamination in the swiss jura,” Environmental Pollution, vol. 86, no. 3,
pp. 315–327, 1994.
[24] P. Goovaerts, Geostatistics for Natural Resources Evaluation. New
York: Oxford University Press, 1997.
[25] M. Schlather, A. Malinowski, P. J. Menck, M. Oesting, and K. Strokorb,
“Analysis, simulation and prediction of multivariate random fields with
package RandomFields,” Journal of Statistical Software, vol. 63, no. 8,
pp. 1–25, 2015. (Online). Available: http://www.jstatsoft.org/v63/i08/
[26] A. G. Journel and C. J. Huijbregts, Mining Geostatistics, reprint ed.
New York: The Blackburn Press, 2003.