Abstract
Globally supported covariance functions are generally associated with dense covariance matrices, meaning severe numerical problems in solution feasibility. These problems can be alleviated by considering methods yielding sparse covariance matrices. Indeed, having many zero entries in the covariance matrix can both greatly reduce computer storage requirements and the number of floating point operations needed in computation. Compactly supported covariance functions considerably reduce the computational burden of kriging, and allow the use of computationally efficient sparse matrix techniques, thus becoming a core aspect in spatial prediction when dealing with massive data sets. However, most of the work done in the context of compactly supported covariance functions has been carried out in the stationary context. This assumption is not generally met in practical and real problems, and there has been a growing recognition of the need for non-stationary spatial covariance functions in a variety of disciplines. In this paper we present a new class of non-stationary, compactly supported spatial covariance functions, which adapts a class of convolution-based flexible models to non-stationary situations. Some particular examples, computational issues, and connections with existing models are considered.
Similar content being viewed by others
References
Arabelos D, Tscherning CC (1996) Collocation with finite covariance functions. Int Geoid Serv Bull 5:17–136
Banerjee S, Carlin BP, Gelfand AE (2004) Hierarchical modeling and analysis for spatial data. Monographs on statistics and applied probability, vol 101. Chapman & Hall, London
Barry R, Pace K (1997) Kriging with large data sets using sparse matrix techniques. Commun Stat Comput Simul 26:619–629
Bernstein SN (1928) Sur les fonctions absolument monotones. Acta Math 52:1–66
Bochner S (1933) Monotone funktionen, Stiltjes integrale und harmonische analyse. Math Ann 108:378–410
Buhmann M (2000) Radial basis functions: the state-of-the-art and new results. Acta Numer 9:1–37
Butt S, Brodlie KW (1993) Preserving positivity using piecewise cubic interpolation. Comput Graph 17:55–64
Calder CA, Cressie N (2007) Some topics in convolution-based spatial modeling. In: Proceedings of the 56th session of the international statistics institute, Lisbon, 22–29 Aug 2007
Christakos G (1984) On the problem of permissible covariance and variogram models. Water Resour Res 20:251–265
Christakos G, Papanicolaou V (2000) Norm-dependent covariance permissibility of weakly homogeneous spatial random fields. Stoch Environ Res Risk Assess 14:1–8
Cressie N (1993) Statistics for spatial data. Wiley series in probability and mathematical statistics: applied probability and statistics. Wiley, New York. Revised reprint of the 1991 edition, A Wiley-Interscience Publication
Diggle PJ, Ribeiro PJ (2006) Model-based geostatistics. Springer series in statistics. Springer, New York
Fuentes M (2002) Spectral methods for nonstationary spatial processes. Biometrika 89:197–210
Fuentes M, Smith R (2001) A new class of nonstationary spatial models. Technical report, Department of Statistics, North Carolina State University
Furrer R, Genton MG, Nychka D (2005) Covariance tapering for interpolation of large spatial datasets. J Comput Graph Stat 15:502–523
Gaspari G, Cohn SE (1999) Construction of correlation functions in two and three dimensions. Quart J R Meteorol Soc 125:723–757
Gneiting T (1999) Correlation functions for atmospheric data analysis. Quart J R Meteorol Soc 125:2449–2464
Gneiting T (2002) Compactly supported correlation functions. J Multivar Anal 83:493–508
Higdon D (2002) Space and space-time modeling using process convolutions. In: Anderson C et al (eds) Quantitative methods for current environmental issues. Springer, London, pp 37–54
Higdon D, Swall J, Kern J (1999) Non-stationary spatial modeling. In: Bernardo JM, Berger JO, Dawid AP, Smith AFM (eds) Bayesian statistics, vol 6. Oxford University Press, Oxford, pp 761–768
Holland DM, Cox WM, Scheffe R, Cimorelli AJ, Nychka D, Hopke PK (2003) Spatial prediction of air quality data. Environ Manag (8):31–35
Hollingsworth A, Lönnberg P (1986) The statistical structure of short-range forecast errors as determined from radiosonde data. Part I: the wind field. Tellus 38:111–136
Hon YC, Zhou X (2000) A comparison on using various radial basis functions for options pricing. Int J Appl Sci Comput 7:29–47
Hussain MZ, Sarfraz M (2008) Positivity-preserving interpolation of positive data by rational cubics. J Comput Appl Math 218:446–458
Jun M, Stein ML (2008) Nonstationary covariance models for global data. Ann Appl Stat 2:1271–1289
Lang T, Plagemann C, Burgard W (2007) Adaptive non-stationary kernel regression for terrain modelling. In: Proceedings of robotics: science and systems (RSS), Atlanta
Lönnberg P, Hollingsworth A (1986) The statistical structure of short-range forecast errors determined from radiosonde data. Part II: the covariance of height and wind errors. Tellus 38:137–161
Moreaux G (2008) Compactly supported radial covariance functions. J Geodesy 82:431–443
Moreaux C, Tscherning C, Sanso F (1999) Approximation of harmonic covariance functions on the sphere by non-harmonic locally supported functions. J Geodesy 73:555–567
Morse BS, Yoo TS, Rheingans P, Chen DT, Subramanian KR (2001) Interpolating implicit surfaces from scattered surface data using compactly supported radial basis functions. In: Proceedings of the international conference on shape modeling and applications, Genova, Italy, pp 89–98
Nychka D, Wikle C, Royle A (2002) Multiresolution models for nonstationary spatial covariance functions. Stat Model 2:299–314
Ong BH, Wong HC (1996) A C1 positivity preserving scattered data interpolation scheme. In: Fontanella F, Jetter K, Laurent P-J (eds) Advanced topics in multivariate approximation, vol 8. World Scientific Publishing, Singapore, pp 259–274
Paciorek C (2003) Nonstationary Gaussian processes for regression and spatial modelling. Unpublished PhD dissertation, Department of Statistics, Carnegie Mellon University
Paciorek C, Schervish M (2004) Nonstationary covariance functions for Gaussian process regression. In: Thrun S, Saul L, Schlkopf B (eds) Advances in neural information processing systems 16. MIT Press, Cambridge, pp 273–280
Paciorek CJ, Schervish MJ (2006) Spatial modelling using a new class of nonstationary covariance functions. Environmetrics 17:483–506
Piah ARM, Saaban A, Majid AA (2006) Range restricted positivity-preserving scattered data interpolation. J Fundam Sci 2:63–75
Pintore A, Holmes CC (2004) Non-stationary covariance functions via spatially adaptive spectra. J Am Stat Assoc (accepted)
Plagemann C, Kersting K, Burgard W (2008) Nonstationary Gaussian process regression using point estimates of local smoothness. In: Proceedings of the European conference on machine learning and knowledge discovery in databases. Lecture notes in computer science, vol 5212. Springer, Heidelberg, pp 204–219
Porcu E, Gregori P, Mateu J (2006) Nonseparable stationary anisotropic space time covariance functions. Stoch Environ Res Risk Assess 21:113–122
Porcu E, Mateu J, Saura F (2008) New classes of covariance and spectral density functions for spatio-temporal modelling. Stoch Environ Res Risk Assess 22:65–79
Porcu E, Mateu J, Christakos G (2010) Quasi-arithmetic means of covariance functions with potential applications to space-time data. J Multivar Anal 100(8):1830–1844
Priestley MB (1965) Evolutionary spectra and nonstationary processes. J R Stat Soc B 27:204–237
Rygaard-Hjalsted C, Constable CG, Parker RL (1997) The influence of correlated crustal signals in modelling the main geomagnetic field. Geophys J Int 130:717–726
Sansò F, Schuh WD (1987) Finite covariance functions. Bull Geod 61:331–347
Sarfraz M (2000) Piecewise rational interpolation preserving positive data. In: Proceedings of the international conference on imaging science, systems, and technology, Las Vegas, pp 103–109
Schaback R (1995) Error estimates and condition numbers for radial basis function interpolation. Adv Comput Math 3:251–264
Schaback R, Wendland H (1993) Special cases of compactly supported radial basis functions. Manuscript, Göttingen
Schmidt JW, Hess W (1988) Positivity of cubic polynomials on intervals and positive spline interpolation. BIT 28:340–352
Schoenberg IJ (1938) Metric spaces and completely monotone functions. Ann Math 39:811–841
Schreiner M (1997) Locally supported kernels for spherical spline interpolation. J Approx Theory 89:172–194
Seeger M (2004) Gaussian processes for machine learning. Int J Neural Syst 14:69–106
Stein ML (1999) Interpolation of spatial data: some theory for kriging. Springer, New York
Stein ML (2005) Space-time covariance functions. J Am Stat Assoc 100:310–321
Tóth G, Völgyesi L (2007) Local gravity field modeling using surface gravity gradient measurements. In: Tregoning P, Rizos C (eds) Dynamic planet monitoring and understanding a dynamic planet with geodetic and oceanographic tools. IAGSymposia 130. Springer, Berlin, pp 424–429
Wachowiak MP, Wang X, Fenster A, Peters TM (2004) Compact support radial basis functions for soft tissue deformation. In: Proceedings of IEEE international symposium on biomedical imaging, Arlington, pp 1259–1262
Wendland H (1995) Piecewise polynomial, positive definite and compactly supported radial functions of minimal degree. Adv Comput Math 4:389–396
Wong SM, Hon Y, Li TS (1999) Radial basis functions with compact support and multizone decomposition: applications to environmental modelling. Bound Elem Technol 13:355–364
Wu Z (1995) Compactly supported positive definite radial functions. Adv Comput Math 4:283–292
Wu J, Zhang X, Peng L (2010) Positive approximation and interpolation using compactly supported radial basis functions. Math Probl Eng, Art no 964528
Acknowledgements
The authors are indebted to two referees and an AE for useful discussions and comments that improved an earlier version of the manuscript. We also would like to thank F. Rodríguez-Cortés for his help with programming. The work was partially funded by grant MTM2010-14961 from the Spanish Ministry of Science and Education.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Mateu, J., Fernández-Avilés, G. & Montero, J.M. On a class of non-stationary, compactly supported spatial covariance functions. Stoch Environ Res Risk Assess 27, 297–309 (2013). https://doi.org/10.1007/s00477-011-0510-8
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00477-011-0510-8