Skip to main content

Interpolation of spatial data – A stochastic or a deterministic problem?

  • M. SCHEUERER (a1), R. SCHABACK (a2) and M. SCHLATHER (a3)

Interpolation of spatial data is a very general mathematical problem with various applications. In geostatistics, it is assumed that the underlying structure of the data is a stochastic process which leads to an interpolation procedure known as kriging. This method is mathematically equivalent to kernel interpolation, a method used in numerical analysis for the same problem, but derived under completely different modelling assumptions. In this paper we present the two approaches and discuss their modelling assumptions, notions of optimality and different concepts to quantify the interpolation accuracy. Their relation is much closer than has been appreciated so far, and even results on convergence rates of kernel interpolants can be translated to the geostatistical framework. We sketch different answers obtained in the two fields concerning the issue of kernel misspecification, present some methods for kernel selection and discuss the scope of these methods with a data example from the computer experiments literature.

Hide All
[1]Anderes, E. (2010) On the consistent separation of scale and variance for Gaussian random fields. Ann. Statist. 38 (2), 870893.
[2]Anderson, J. L. (1996) A method for producing and evaluating probabilistic forecasts from ensemble model integrations. J. Clim. 9, 15181530.
[3]Banerjee, S., Gelfand, A. E., Finley, A. O. & Sang, H. (2008) Gaussian predictive process models for large spatial datasets. J. R. Statist. Soc B 70 (4), 825848.
[4]Berlinet, A. & Thomas-Agnan, C. (2004) Reproducing Kernel Hilbert Spaces in Probability and Statistics. Kluwer, Berlin, Germany.
[5]Caragea, P. & Smith, R. L. (2007) Asymptotic properties of computationally efficient alternative estimators for a class of multivariate normal models. J. Multivariate Anal. 98 (7), 14171440.
[6]Carlson, R. E. & Foley, T. A. (1991) The parameter R 2 in multiquadric interpolation. Comp. Math. Appl. 21, 2942.
[7]Chauvet, P., Pailleux, J. & Chilès, J.-P. (1976) Analyse objective des champs météorologiques par cokrigeage. La Météorologie, 6ième Série 4, 3754.
[8]Chilès, J.-P. (1976) How to adapt kriging to non-classical problems: three case studies. In: Guarascio, M., David, M. & Huijbregts, C. (editors), Advanced Geostatistics in the Mining Industry, D. Reidel, Dordrecht, Holland, pp. 6989.
[9]Chilès, J.-P. & Delfiner, P. (1999) Geostatistics. Modeling Spatial Uncertainty, John Wiley, New York.
[10]Cressie, N. (1989) Geostatistics. Am. Stat. 43 (4), 197202.
[11]Cressie, N. (1993) Statistics for Spatial Data (rev. ed. edition), Wiley, New York.
[12]Cressie, N. & Johannesson, G. (2008) Fixed rank kriging for very large spatial data sets. J. R. Statist. Soc B 70 (1), 209226.
[13]Dahlhaus, R. & Künsch, H. R. (1987) Edge effects and efficient parameter estimation for stationary random fields. Biometrika 74 (4), 877882.
[14]de Marchi, S. (2003) On optimal center locations for radial basis function interpolation: Computational aspects. Rend. Sem. Mat. Torino 61 (3), 343358.
[15]de Marchi, S. & Schaback, R. (2010) Stability of kernel-based interpolation. Adv. Comput. Math. 32, 155161.
[16]de Marchi, S., Schaback, R. & Wendland, H. (2005) Near-optimal data-independent point locations for radial basis functions. Adv. Comput. Math. 23 (3), 317330.
[17]Diggle, P. J. & Ribeiro, P. J. (2007) Model-Based Geostatistics, Springer, Berlin, Germany.
[18]Duchon, J. (1976) Interpolation des fonctions de deux variables suivant le principe de la flexion des plaques minces. Adv. RAIRO Anal. Num. 10, 512.
[19]Duchon, J. (1977) Splines minimizing rotation invariant seminorms in Sobolev spaces. In: Schempp, W. & Zeller, K. (editors), Constructive Theory of Functions of Several Variables, Springer-Verlag, Berlin, Germay, pp. 85100.
[20]Duchon, J. (1978) Sur l'erreur d'interpolation des fonctions de plusieurs variables par les Dm-splines. Adv. RAIRO Anal. Num. 12, 325334.
[21]Eidsvik, J., Finley, S., Banerjee, S. & Rue, H. (2010) Approximate Bayesian Inference for Large Spatial Datasets using Predictive Process Models. Technical Report 9, Department of Mathematical Sciences, Norwegian University of Science and Technology, Norway.
[22]Evans, L. C. (2002) Partial Differential Equations, American Mathematical Society, Providence, RI.
[23]Fasshauer, G. E. (2007) Meshfree Approximation Methods with Matlab, World Scientific, Singapore.
[24]Fasshauer, G. E. & Zhang, J. G. (2007) On choosing “optimal” shape parameters for RBF approximation. Numer. Algorithms 45, 345368.
[25]Foley, T. A. (1987) Interpolation and approximation of 3-D and 4-D scattered data. Comput. Math. Appl. 13, 711740.
[26]Fornberg, B., Driscoll, T. A., Wright, G. & Charles, R. (2002) Observations on the behaviour of radial basis function approximations near boundaries. Comput. Math. Appl. 43, 473490.
[27]Fornberg, B. & Wright, G. (2004) Stable computation of multiquadric interpolants for all values of the shape parameter. Comput. Math. Appl. 47, 497523.
[28]Franke, R. (1982) Scattered data interpolation: Tests of some methods. Math. Comput. 38, 181200.
[29]Fuentes, M. (2008) Approximate likelihood for large irregular spaced spatial data. J. Am. Stat. Assoc. 102 (477), 321331.
[30]Furrer, R., Genton, M. G. & Nychka, D. (2006) Covariance tapering for interpolation of large spatial datasets. J. Comput. Graph. Stat. 15 (3), 502523.
[31]Gneiting, T. & Schlather, M. (2004) Stochastic models that separate fractal dimension and the Hurst effect. SIAM Rev. 46 (2), 269282.
[32]Guyon, X. (1982) Parameter estimation for a stationary process on a d-dimensional lattice. Biometrika 69 (1), 95105.
[33]Handcock, M. S. & Stein, M. L. (1993) A Bayesian analysis of kriging. Technometrics 35 (4), 403410.
[34]Handcock, M. S. & Wallis, J. R. (1994) An approach to statistical spatial-temporal modeling of meteorological fields (with discussion). J. Am. Stat. Assoc. 89 (7), 368390.
[35]Hardy, R. L. (1971) Multiquadric equations of topography and other irregular surfaces. J. Geophys. Res. 76, 19051915.
[36]Harville, D. A. (1974) Bayesian inference for variance components using only error contrasts. Biometrika 61, 383385.
[37]Heyde, C. C. (1997) Quasi-Likelihood and Its Application, Springer, New York.
[38]Ibragimov, I. A. & Rozanov, Y. A. (1978) Gaussian Random Processes, Aries, A. B. (trans.), Springer, New York.
[39]Iske, A. (2000) Optimal Distributions of Centers for Radial Basis Function Methods. Technical Report M0004, Technische Universität München, Munich, Germany.
[40]Joseph, V. R., Hung, Y. & Sudjianto, A. (2008) Blind kriging: A new method for developing metamodels. J. Mech. Des. 130 (3), 18.
[41]Journel, A. G. (1982) The indicator approach to estimation of spatial distributions. In Proceedings of the 17th APCOM International Symposium, New York, pp. 793806.
[42]Kimeldorf, G. S. & Wahba, G. (1970) A correspondence between Bayesian estimation on stochastic processes and smoothing by splines. Ann. Math. Stat. 41 (2), 495502.
[43]Kitanidis, P. K. (1983) Statistical estimation of polynomial generalized covariance functions and hydrologic applications. Water Resour. Res. 19 (4), 909921.
[44]Kitanidis, P. K. (1997) Introduction to Geostatistics: Applications in Hydrology, Cambridge University Press, New York.
[45]Lajaunie, C. & Béjaoui, R. (1991) Sur le Krigeage des Fonctions Complexes. Technical Report N-23/91/G, Centre de Géostatistique, Ecole des Mines de Paris, Fontainebleau.
[46]Lim, S. C. & Teo, L. P. (2010) Analytic and asymptotic properties of multivariate generalized Linniks probability densities. J. Fourier Anal. Appl. 16, 715747.
[47]Lindgren, F., Rue, H. & Lindström, J. (2011) An explicit link between Gaussian fields and Gaussian Markov random fields: The stochastic partial differential equation approach. J. R. Stat. Soc B 73 (4), 423498.
[48]Madych, W. R. & Nelson, S. A. (1988) Multivariate interpolation and conditionally positive definite functions. Approx. Theory Appl. 4, 7789.
[49]Mardia, K. V. & Marshall, R. J. (1984) Maximum likelihood estimation of models for residual covariance in spatial statistics. Biometrika 71, 135146.
[50]Matérn, B. (1986) Spatial Variation, 2nd ed., Lecture Notes in Statistics, Vol. 36, Springer-Verlag, Berlin, Germany.
[51]Matheron, G. (1971) The Theory of Regionalized Variables and its Applications. Technical Report, Cahiers du Centre de Morphologie Mathḿatique de Fontainebleau, Ecole des Mines de Paris.
[52]Matheron, G. (1973) The intrinsic random functions and their applications. Adv. Appl. Prob. 5, 439468.
[53]Matheron, G. (1973) Le Krigeage Disjonctive. Technical Report N-360, Centre de Géostatistique, Ecole des Mines de Paris.
[54]Matheron, G. (1976) A simple substitute for conditional expectation: The disjunctive kriging. In: Guarascio, M., David, M. & Huijbregts, C. (editors), Advanced Geostatistics in the Mining Industry, Reidel, Dordrecht, Netherland, pp. 221236.
[55]Micchelli, C. A. (1986) Interpolation of scattered data: Distance matrices and conditionally positive definite functions. Constr. Approx. 2, 1122.
[56]Morris, M. D., Mitchell, T. J. & Ylvisaker, D. (1993) Bayesian design and analysis of computer experiments: Use of derivatives in surface prediction. Technometrics 35 (3), 243255.
[57]Myers, D. E. (1992) Kriging, cokriging, radial basis functions and the role of positive definiteness. Comput. Math. Appl. 24 (12), 139148.
[58]Narcowich, F. J., Ward, J. D. & Wendland, H. (2006) Sobolev error estimates and a Bernstein inequality for scattered data interpolation via radial basis functions. Constr. Approx. 24, 175186.
[59]Omre, H. & Halvorsen, K. B. (1989) The Bayesian bridge between simple and universal kriging. Math. Geol. 21 (7), 767786.
[60]Putter, H. & Young, G. A. (2001) On the effect of covariance function estimation on the accuracy of kriging predictors. Bernoulli 7 (3), 421438.
[61]R Development Core Team. (2011) R: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing, Vienna, Austria.
[62]Rasmussen, C. E. & Williams, C. K. I. (2006) Gaussian Processes for Machine Learning, MIT Press, Boston, MA.
[63]Rippa, S. (1999) An algorithm for selecting a good value for the parameter c in radial basis function interpolation. Adv. Comput. Math. 11, 193210.
[64]Ritter, K. (2000) Average-Case Analysis of Numerical Problems, Lecture Notes in Mathematics, No. 1733, Springer, New York.
[65]Schaback, R. (1993) Comparison of radial basis function interpolants. In: Jetter, K. & Utreras, F. (editors), Multivariate Approximation: From CAGD to Wavelets, World Scientific, London, pp. 293305.
[66]Schaback, R. (1995) Error estimates and condition numbers for radial basis function interpolation. Adv. Comput. Math. 3, 251264.
[67]Schaback, R. (1997) Native Hilbert spaces for radial basis functions I. In: New Developments in Approximation Theory, International Series of Numerical Mathematics, No. 132, Birkhauser Verlag, Berlin, Germany, pp. 255282.
[68]Schaback, R. (2011) The missing Wendland functions. Adv. Comput. Math. 34, 6781.
[69]Schaback, R. (2011) Kernel-Based Meshless Methods. Technical report, Georg-August-Universität Göttingen, Göttingen, Germany.
[70]Schaback, R. & Wendland, H. (2006) Kernel techniques: From machine learning to meshless methods. Acta Numer. 15, 543639.
[71]Scheuerer, M. (2010) Regularity of the sample paths of a general second-order random field. Stoch. Proc. Appl. 120, 18791897.
[72]Scheuerer, M. (2011) An alternative procedure for selecting a good value for the parameter c in RBF-interpolation. Adv. Comput. Math. 34 (1), 105126.
[73]Schlather, M. (2001) RandomFields: Contributed extension package to R for the simulation of Gaussian and max-stable random fields. URL:
[74]Seeger, M. (2004) Gaussian processes for machine learning. Int. J. Neural Syst. 14, 138.
[75]Stein, M. L. (1988) Asymptotically efficient prediction of a random field with a misspecified covariance function. Ann. Stat. 16, 5563.
[76]Stein, M. L. (1990) A comparison of generalized cross validation and modified maximum likelihood for estimating the parameters of a stochastic process. Ann. Stat. 18 (3), 11391157.
[77]Stein, M. L. (1999) Interpolation of Spatial Data, Springer, New York.
[78]Stein, M. L. (2004) Equivalence of Gaussian measures for some nonstationary random fields. J. Stat. Plann. Inference 123, 111.
[79]Stein, M. L., Chi, Z. & Welty, L. J. (2004) Approximating likelihoods for large spatial data sets. J. R. Stat. Soc B 66 (2), 275296.
[80]van der Vaart, A. W. & van Zanten, J. H. (2011) Information rates of nonparametric Gaussian process methods. J. Mach. Learn. Res. 12, 20952119.
[81]Vecchia, A. V. (1988) Estimation and model identification for continuous spatial processes. J. R. Stat. Soc B 50, 297312.
[82]Wahba, G. (1985) A comparison of gcv and gml for choosing the smoothing parameter in the generalized spline smoothing problem. Ann. Stat. 13 (1), 13781402.
[83]Wahba, G. (1990) Spline Models for Observational Data, SIAM, Philadelphia, PA.
[84]Wendland, H. (1995) Piecewise polynomial, positive definite and compactly supported radial functions of minimal degree. Adv. Comput. Math. 4, 389396.
[85]Wendland, H. (2005) Scattered Data Approximation, Cambridge Monographs on Applied and Computational Mathematics, Cambridge University Press, Cambridge, UK.
[86]Wendland, H. & Rieger, C. (2005) Approximate interpolation with applications to selecting smoothing parameters. Numer. Math. 101, 729748.
[87]Wu, Z. (1992) Hermite-Birkhoff interpolation of scattered data by radial basis functions. Approx. Theory Appl. 8 (2), 110.
[88]Zhang, H. (2004) Inconsistent estimation and asymptotically equivalent interpolations in model-based geostatistics. J. Am. Stat. Assoc. 99, 250261.
Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

European Journal of Applied Mathematics
  • ISSN: 0956-7925
  • EISSN: 1469-4425
  • URL: /core/journals/european-journal-of-applied-mathematics
Please enter your name
Please enter a valid email address
Who would you like to send this to? *



Altmetric attention score

Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed