Abstract
Kriging is an interpolation method that consists of finding a predictor linear function of the observations, minimizing the mean squared prediction error or kriging variance. Under multivariate normality assumptions, the given predictor is the best linear unbiased predictor, but if the underlying distribution is not normal, the estimator shall not be unbiased and shall be vulnerable to outliers. In the spatial context, it is not only the presence of outliers that may spoil the predictions, but also the boundary sites, usually corners, that tend to have high leverage. Therefore, kriging predictions are very sensitive on these corners, giving rise to values extremely vulnerable to small changes in the data. To overcome this situation, a spatial linear mixed model is proposed, deriving a bounded influence estimator of the location parameters. To illustrate the results, an application to Davis topographic data is presented.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Aitkin, M. and Francis B. J. (1995). Fitting overdispersed generalized linear models by nonparametric maximum likelihood. GLIM Newsletter, 25, 37–45.
Cressie, N. A. C. (1991). Statistics for Spatial Data. Wiley Series in Probability and Mathematical Statistics.
Cressie, N. and Hawkins, D. M. (1980). Robust estimation of the variogram. Mathematical Geology, 12 115–125.
Davis, J. C. (1986). Statistics and Data Analysis in Geology. 2nd ed. John Wiley and Sons, New York.
Dempster, A. P., Laird, N. M. and Rubin, D. B. (1977). Maximum likelihood from incomplete data via the EM-algorithm. Journal of the Royal Statistical Society,Series B,39 1–38.
Hawkins, D. M. and Cressie, N. A. C. (1984). Robust kriging: A proposal. Mathematical Geology, 16, 3–18.
Heckman, J. J. and Singer, B. (1984). A method for minimizing the impact of distributional assumptions in econometric models of duration. Econometrica, 52 271–320.
Hoaglin, D. C., Mosteller, F. and Tukey, J. W. (1985). Exploring Data Tables, Trends,and Shapes. Wiley Series in Probability and Statistics.
Laird, N. M. (1978). Nonparametric maximum likelihood estimation of a mixing distribution. Journal of the American Statistical Association, 73, 805–811.
Mallows, C. L. (1975). On some topics in Robustness. Technical memorandum, Bell Telephone Laboratories, Murray Hill, NJ.
McLachlan, G. J. and Basford, K. E. (1988). Mixture Models. Inference and Applications to clustering. Marcel Dekker.
Militino, A. F. (1996). M-estimation of the drift coefficients of a spatial linear model. Mathematical Geology, in press.
Militino, A. F. and Ugarte, M. D. (1996). Robust estimation of the location parameters of a spatial linear model via random effects, in Statistical Modelling, 275–282, edited by G.M. Marchetti, A. Forcina, R. Hatzinger and G. Galmacci. Published in Graphos, Citta’ di Castello, Italy.
Ripley, B. D. (1981). Spatial Statistics. Wiley Series in Probability and Mathematical Statistics.
Rousseeuw, P. J. and van Zomeren, B. C. (1990). Unmasking multivariate outliers and leverage points (with discussion). Journal of the American Statistical Association, 85 633–651.
Simpson, D. G., Ruppert, D. and Carroll, R. J. (1992). On one-step GM estimates and stability of inferences in linear regression. Journal of the American Statistical Association, 87,439–450.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1997 Springer Science+Business Media New York
About this paper
Cite this paper
Militino, A.F., Ugarte, M.D. (1997). Bounded Influence Estimation in a Spatial Linear Mixed Model. In: Gregoire, T.G., Brillinger, D.R., Diggle, P.J., Russek-Cohen, E., Warren, W.G., Wolfinger, R.D. (eds) Modelling Longitudinal and Spatially Correlated Data. Lecture Notes in Statistics, vol 122. Springer, New York, NY. https://doi.org/10.1007/978-1-4612-0699-6_18
Download citation
DOI: https://doi.org/10.1007/978-1-4612-0699-6_18
Publisher Name: Springer, New York, NY
Print ISBN: 978-0-387-98216-8
Online ISBN: 978-1-4612-0699-6
eBook Packages: Springer Book Archive