Advertisement

Multivariate spatial modelling through a convolution-based skewed process

  • Hamid Zareifard
  • Majid Jafari KhalediEmail author
  • Omar Dahdouh
Original Paper
  • 187 Downloads

Abstract

In some statistical issues, several continuous spatial outcomes are simultaneously measured at each sampling location. In such circumstances, it is common to model the data through a multivariate Gaussian model. As the normality assumption is often untenable, this paper proposes a multivariate skewed spatial model which, by virtue of its capacity for capturing skewness, is potentially more flexible than symmetric ones. Specifically, a multivariate version of the Gaussian-log Gaussian convolution process is developed. The resulting covariance for the multivariate process is in general nonseparable. We also discuss the other properties of the induced covariance function. Furthermore, Markov chain Monte Carlo methods are used to make Bayesian inferences. The performance of the method is investigated through simulation experiments and by analyzing a real soil pollution dataset obtained from Golestan province, North of Iran.

Keywords

Multivariate spatial data Non-Gaussian modelling Bayesian analysis Convolution Cross-dependency structure 

Notes

Acknowledgements

We thank two anonymous reviewers, the associate editor, and the editor for their constructive comments. Their very much precise comments have substantially improved the manuscript.

References

  1. Alegría A, Caro C, Bevilacqua M, Porcu E, Clarke J (2017) Estimating covariance functions of multivariate skew-Gaussian random fields on the sphere. Spat Stat 22:388–402CrossRefGoogle Scholar
  2. Allard D, Naveau P (2007) A new spatial skew-normal random field model. Commun Stat Theory Methods 36(9):1821–1834CrossRefGoogle Scholar
  3. Alvarez I, Niemi J, Simpson M (2014) Bayesian inference for a covariance matrix. arXiv preprint arXiv:1408.4050Google Scholar
  4. Bandyopadhyay S, Rao SS (2017) A test for stationarity for irregularly spaced spatial data. J R Stat Soc Ser B (Stat Methodol) 79(1):95–123CrossRefGoogle Scholar
  5. Bartoletti S, Loperfido N (2010) Modelling air pollution data by the skew-normal distribution. Stoch Environ Res Risk Assess 24(4):513–517CrossRefGoogle Scholar
  6. Bolin D (2014) Spatial Matérn fields driven by non-Gaussian noise. Scand J Stat 41:557–579CrossRefGoogle Scholar
  7. Bourgault G, Marcotte D (1991) Multivariable variogram and its application to the linear model of coregionalization. Math Geol 23(7):899–928CrossRefGoogle Scholar
  8. Bárdossy A, Li J (2008) Geostatistical interpolation using copulas. Water Resour Res 44:W07412.  https://doi.org/10.1029/2007WR006115 CrossRefGoogle Scholar
  9. Cressie N, Zammit-Mangion A (2016) Multivariate spatial covariance models: a conditional approach. Biometrika 103(4):915–935CrossRefGoogle Scholar
  10. De Oliveira V (2003) A note on the correlation structure of transformed Gaussian random fields. Aust N Z J Stat 45(3):353–366CrossRefGoogle Scholar
  11. De Oliveira V (2004) A simple model for spatial rainfall fields. Stoch Environ Res Risk Assess 18(2):131–140CrossRefGoogle Scholar
  12. De Oliveira V, Kedem B, Short DA (1997) Bayesian prediction of transformed Gaussian random fields. J Am Stat Assoc 92:1422–1433Google Scholar
  13. Fonseca TC, Steel MF (2011) Non-Gaussian spatiotemporal modelling through scale mixing. Biometrika 98(4):761–774CrossRefGoogle Scholar
  14. Fouedjio F (2018) A fully non-stationary linear coregionalization model for multivariate random fields. Stoch Environ Res Risk Assess 32(6):1699–1721CrossRefGoogle Scholar
  15. Garcia-Papani F, Uribe-Opazo MA, Leiva V, Aykroyd RG (2017) BirnbaumSaunders spatial modelling and diagnostics applied to agricultural engineering data. Stoch Environ Res Risk Assess 31(1):105–124CrossRefGoogle Scholar
  16. Garcia-Papani F, Leiva V, Uribe-Opazo MA, Aykroyd RG (2018) Birnbaum–Saunders spatial regression models: diagnostics and application to chemical data. Chemom Intell Lab Syst 177:114–128CrossRefGoogle Scholar
  17. Gelfand AE, Diggle P, Guttorp P, Fuentes M (eds) (2010) Handbook of spatial statistics. CRC Press, Boca RatonGoogle Scholar
  18. Gelman A, Rubin DB (1992) Inference from iterative simulation using multiple sequences. Stat Sci 7(4):457–472CrossRefGoogle Scholar
  19. Genton MG, Kleiber W (2015) Cross-covariance functions for multivariate geostatistics. Stat Sci 30(2):147–163CrossRefGoogle Scholar
  20. Genton MG, Zhang H (2012) Identifiability problems in some non-Gaussian spatial random fields. Chil J Stat 3(2):171–179Google Scholar
  21. Gneiting T, Kleiber W, Schlather M (2010) Matérn cross-covariance functions for multivariate random fields. J Am Stat Assoc 105(491):1167–1177CrossRefGoogle Scholar
  22. Goulard M, Voltz M (1992) Linear coregionalization model: tools for estimation and choice of cross-variogram matrix. Math Geol 24(3):269–286CrossRefGoogle Scholar
  23. Grzebyk M, Wackernagel H (1994) Multivariate analysis and spatial/temporal scales: real and complex models. In: Proceedings of the XVIIth international biometrics conference, vol 1, pp 19–33Google Scholar
  24. Ip RH, Li WK (2016) Matérn cross-covariance functions for bivariate spatio-temporal random fields. Spat Stat 17:22–37CrossRefGoogle Scholar
  25. Ip RH, Li WK (2017) A class of valid Matérn cross-covariance functions for multivariate spatio-temporal random fields. Stat Probab Lett 130:115–119CrossRefGoogle Scholar
  26. Kass RE, Raftery AE (1995) Bayes factors. J Am Stat Assoc 90(430):773–795CrossRefGoogle Scholar
  27. Kazianka H, Pilz J (2009) A corrected criterion for selecting the optimum number of principal components. Austrian J Stat 38(3):135–150CrossRefGoogle Scholar
  28. Kazianka H, Pilz J (2010) Copula-based geostatistical modelling of continuous and discrete data including covariates. Stoch Environ Res Risk Assess 24(5):661–673CrossRefGoogle Scholar
  29. Kim HM, Mallik BK (2004) A Bayesian prediction using the skew-Gaussian distribution. J Stat Planing Inference 120:85–101CrossRefGoogle Scholar
  30. Krupskii P, Genton MG (2017) Factor copula models for data with spatio-temporal dependence. Spat Stat 22:180–195CrossRefGoogle Scholar
  31. Marchenko YV, Genton MG (2010) Multivariate log-skew-elliptical distributions with applications to precipitation data. Environmetrics 21(3–4):318–340Google Scholar
  32. Mastrantonio G, Maruotti A, JonaLasinio G (2015) Bayesian hidden Markov modelling using circularlinear general projected normal distribution. Environmetrics 26(2):145–158CrossRefGoogle Scholar
  33. Meng XL, Rubin DB (1991) Using EM to obtain asymptotic variance-covariance matrices: the SEM algorithm. J Am Stat Assoc 86(416):899–909CrossRefGoogle Scholar
  34. Newton MA, Raftery AE (1994) Approximate Bayesian inference with the weighted likelihood bootstrap. J R Stat Soc Ser B (Methodol) 56(1):3–26Google Scholar
  35. North GR, Wang J, Genton MG (2011) Correlation models for temperature fields. J Clim 24(22):5850–5862CrossRefGoogle Scholar
  36. Palacios MB, Steel MFJ (2006) Non-gaussian Bayesian geostatistical modelling. J Am Stat Assoc 101(474):604–618CrossRefGoogle Scholar
  37. Pilz J, Spöck G (2008) Why do we need and how should we implement Bayesian kriging methods. Stoch Environ Res Risk Assess 22(5):621–632CrossRefGoogle Scholar
  38. Pilz J, Kazianka H, Spöck G (2008) Interoperability-spatial interpolation and automated mapping. naGoogle Scholar
  39. Rimstad K, Omre H (2014) Skew-Gaussian random fields. Spat Stat 10:43–62CrossRefGoogle Scholar
  40. Rodríguez A, Dunson DB, Gelfand AE (2010) Latent stick-breaking processes. J Am Stat Assoc 105(490):647–659CrossRefGoogle Scholar
  41. Rue H, Martino S, Chopin N (2009) Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations. J R Stat Soc Ser B (Stat Methodol) 71(2):319–392CrossRefGoogle Scholar
  42. Schmidt AM, Gelfand AE (2003) A Bayesian coregionalization approach for multivariate pollutant data. J Geophys Res Atmos 108:1–9Google Scholar
  43. Schmidt AM, Gonçalves KC, Velozo PL (2017) Spatiotemporal models for skewed processes. Environmetrics 28(6):e2411CrossRefGoogle Scholar
  44. Spöck G, Kazianka H, Pilz J (2009) Bayesian trans-Gaussian kriging with log-log transformed skew data. In: Interfacing geostatistics and GIS. Springer, Berlin, pp 29–43Google Scholar
  45. Tan MT, Tian GL, Ng KW (2010) Bayesian missing data problems: EM, data augmentation and noniterative computation. Chapman and Hall/CRC, Boca RatonGoogle Scholar
  46. Tokuda T, Goodrich B, Van Mechelen I, Gelman A, Tuerlinckx F (2011) Visualizing distributions of covariance matrices. Columbia University, New York, USA, Technical Report, vol 18, p 18Google Scholar
  47. Vargas-Guzmán JA, Warrick AW, Myers DE (2002) Coregionalization by linear combination of nonorthogonal components. Math Geol 34(4):405–419CrossRefGoogle Scholar
  48. Ver Hoef JM, Barry RP (1998) Constructing and fitting models for cokriging and multivariable spatial prediction. J Stat Planning Inference 69(2):275–294CrossRefGoogle Scholar
  49. Ver Hoef JM, Cressie N, Barry RP (2004) Flexible spatial models for kriging and cokriging using moving averages and the Fast Fourier Transform (FFT). J Comput Graph Stat 13(2):265–282CrossRefGoogle Scholar
  50. Wackernagel H (2003) Multivariate geostatistics: an introduction with applications, 3rd edn. Springer, BerlinCrossRefGoogle Scholar
  51. Wallin J, Bolin D (2015) Geostatistical modelling using non-Gaussian Matrn fields. Scand J Stat 42(3):872890CrossRefGoogle Scholar
  52. Xu G, Genton MG (2017) Tukey g-and-h random fields. J Am Stat Assoc 112(519):1236–1249CrossRefGoogle Scholar
  53. Yan Y, Genton MG (2018) Non-Gaussian autoregressive processes with Tukey g-and-h transformations. Environmetrics.  https://doi.org/10.1002/env.2503 Google Scholar
  54. Zareifard H, Jafari Khaledi M (2013) Non-Gaussian modelling of spatial data using scale mixing of a unified skew Gaussian process. J Multivar Anal 114:16–28CrossRefGoogle Scholar
  55. Zareifard H, Rue H, Jafari Khaledi M, Lindgren F (2016) A skew Gaussian decomposable graphical model. J Multivar Anal 145:58–72CrossRefGoogle Scholar
  56. Zareifard H, Jafari Khaledi M, Rivaz F, Vahidi-Asl MQ (2018) Modelling skewed spatial data using a convolution of Gaussian and log-Gaussian processes. Bayesian Anal 13(2):531–557CrossRefGoogle Scholar
  57. Zhang H, El-Shaarawi A (2010) On spatial skew-Gaussian processes and applications. Environmetrics 21(1):33–47Google Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Department of StatisticsJahrom UniversityJahromIran
  2. 2.Department of StatisticsTarbiat Modares UniversityTehranIran

Personalised recommendations