Advertisement

Non-Stationary Dependence Structures for Spatial Extremes

  • Raphaël HuserEmail author
  • Marc G. Genton
Article

Abstract

Max-stable processes are natural models for spatial extremes because they provide suitable asymptotic approximations to the distribution of maxima of random fields. In the recent past, several parametric families of stationary max-stable models have been developed, and fitted to various types of data. However, a recurrent problem is the modeling of non-stationarity. In this paper, we develop non-stationary max-stable dependence structures in which covariates can be easily incorporated. Inference is performed using pairwise likelihoods, and its performance is assessed by an extensive simulation study based on a non-stationary locally isotropic extremal t model. Evidence that unknown parameters are well estimated is provided, and estimation of spatial return level curves is discussed. The methodology is demonstrated with temperature maxima recorded over a complex topography. Models are shown to satisfactorily capture extremal dependence.

Keywords

Covariate Extremal t model Extreme event Max-stable process Non-stationarity 

Supplementary material

13253_2016_247_MOESM1_ESM.pdf (250 kb)
Supplementary material 1 (pdf 250 KB)

References

  1. Anderes, E. B. and Stein, M. L. (2008) Estimating Deformations of Isotropic Gaussian Random Fields on the Plane. Annals of Statistics 36(2), 719–741.Google Scholar
  2. ——– (2011) Local Likelihood Estimation for Nonstationary Random Fields. Journal of Multivariate Analysis 102(3), 506–520.Google Scholar
  3. Beirlant, J., Goegebeur, Y., Segers, J. and Teugels, J. (2004) Statistics of Extremes: Theory and Applications. Chichester: Wiley. ISBN 9780471976479.Google Scholar
  4. Blanchet, J. and Davison, A. C. (2011) Spatial Modelling of Extreme Snow Depth. Annals of Applied Statistics 5(3), 1699–1725.Google Scholar
  5. Bornn, L., Shaddick, G. and Zidek, J. V. (2012) Modeling Nonstationary Processes Through Dimension Expansion. Journal of the American Statistical Association 107(497), 281–289.Google Scholar
  6. Brown, B. M. and Resnick, S. I. (1977) Extreme Values of Independent Stochastic Processes. Journal of Applied Probability 14(4), 732–739.Google Scholar
  7. Capéraà, P., Fougères, A.-L. and Genest, C. (1997) A Nonparametric Estimation Procedure for Bivariate Extreme Value Copulas. Biometrika 84(3), 567–577.Google Scholar
  8. de Carvalho, M. (2015) Statistics of Extremes: Challenges and Opportunities. In Extreme Events in Finance, ed. F. Longin. New York: Wiley.Google Scholar
  9. de Carvalho, M. and Davison, A. C. (2014) Spectral Density Ratio Models for Multivariate Extremes. Journal of the American Statistical Association 109(506), 764–776.Google Scholar
  10. Castro, D., de Carvalho, M. and Wadsworth, J. L. (2015) Time-Varying Extreme Value Dependence With Application to Leading European Stock Markets. Submitted.Google Scholar
  11. Castruccio, S. and Genton, M. G. (2016) Compressing an Ensemble with Statistical Models: An Algorithm for Global 3D Spatio-Temporal Temperature. Technometrics To appear.Google Scholar
  12. Castruccio, S., Huser, R. and Genton, M. G. (2016) High-Order Composite Likelihood Inference for Max-Stable Distributions and Processes. Journal of Computational and Graphical Statistics To appear.Google Scholar
  13. Castruccio, S. and Stein, M. L. (2013) Global Space-Time Models for Climate Ensembles. Annals of Applied Statistics 7(3), 1593–1611.Google Scholar
  14. Chavez-Demoulin, V. and Davison, A. C. (2005) Generalized Additive Modelling of Sample Extremes. Journal of the Royal Statistical Society: Series C (Applied Statistics) 54(1), 207–222.Google Scholar
  15. Cooley, D. S., Cisewski, J., Erhardt, R. J., Jeon, S., Mannshardt-Shamseldin, E. C., Omolo, B. O. and Sun, Y. (2012) A Survey of Spatial Extremes: Measuring Spatial Dependence and Modeling Spatial Effects. REVSTAT 10(1), 135–165.Google Scholar
  16. Cooley, D. S., Naveau, P. and Nychka, D. (2007) Bayesian Spatial Modeling of Extreme Precipitation Return Levels. Journal of the American Statistical Association 102(479), 824–840.Google Scholar
  17. Cressie, N. A. C. (1993) Statistics for Spatial Data. New York: Wiley.zbMATHGoogle Scholar
  18. Cressie, N. A. C. and Wikle, C. K. (2011) Statistics for Spatio-Temporal Data. Hoboken: Wiley. ISBN 9780471692744.zbMATHGoogle Scholar
  19. Davis, R. A., Klüppelberg, C. and Steinkohl, C. (2013) Statistical Inference for Max-Stable Processes in Space and Time. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 75(5), 791–819.Google Scholar
  20. Davison, A. C. and Gholamrezaee, M. M. (2012) Geostatistics of Extremes. Proceedings of the Royal Society A: Mathematical, Physical & Engineering Sciences 468(2138), 581–608.Google Scholar
  21. Davison, A. C. and Huser, R. (2015) Statistics of Extremes. Annual Review of Statistics and its Application 2, 203–235.Google Scholar
  22. Davison, A. C., Huser, R. and Thibaud, E. (2013) Geostatistics of Dependent and Asymptotically Independent Extremes. Mathematical Geosciences 45(5), 511–529.Google Scholar
  23. Davison, A. C., Padoan, S. and Ribatet, M. (2012) Statistical Modelling of Spatial Extremes (with Discussion). Statistical Science 27(2), 161–186.Google Scholar
  24. Engelke, S., Malinowski, A., Kabluchko, Z. and Schlather, M. (2015) Estimation of Huesler–Reiss distributions and Brown–Resnick processes. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 77(1), 239–265.Google Scholar
  25. Fuentes, M. (2001) A High Frequency Kriging Approach for Non-Stationary Environmental Processes. Environmetrics 12(5), 469–483.Google Scholar
  26. Genton, M. G., Ma, Y. and Sang, H. (2011) On the Likelihood Function of Gaussian Max-Stable Processes. Biometrika 98(2), 481–488.Google Scholar
  27. de Haan, L. (1984) A Spectral Representation for Max-Stable Processes. Annals of Probability 12(4), 1194–1204.Google Scholar
  28. de Haan, L. and Ferreira, A. (2006) Extreme Value Theory: An Introduction. New York: Springer. ISBN 9780387239460.Google Scholar
  29. Huser, R. and Davison, A. C. (2013) Composite Likelihood Estimation for the Brown–Resnick Process. Biometrika 100(2), 511–518.Google Scholar
  30. ——– (2014) Space-Time Modelling of Extreme Events. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 76(2), 439–461.Google Scholar
  31. Huser, R., Davison, A. C. and Genton, M. G. (2016) Likelihood Estimators for Multivariate Extremes. Extremes 19(1), 79–103.Google Scholar
  32. Hüsler, J. and Reiss, R.-D. (1989) Maxima of Normal Random Vectors: Between Independence and Complete Dependence. Statistics & Probability Letters 7(4), 283–286.Google Scholar
  33. Jeon, S. and Smith, R. L. (2012) Dependence Structure of Spatial Extremes Using Threshold Approach. arXiv:1209.6344v1.
  34. Jun, M. and Stein, M. L. (2007) An Approach to Producing Space-Time Covariance Functions on Spheres. Technometrics 49(4), 468–479.Google Scholar
  35. ——– (2008) Nonstationary Covariance Models for Global Data. Annals of Applied Statistics 2(4), 1271–1289.Google Scholar
  36. Kabluchko, Z. and Schlather, M. (2010) Ergodic Properties of Max-Infinitely Divisible Processes. Stochastic Processes and their Applications 120(3), 281–295.Google Scholar
  37. Kabluchko, Z., Schlather, M. and de Haan, L. (2009) Stationary Max-Stable Fields Associated to Negative Definite Functions. Annals of Probability 37(5), 2042–2065.Google Scholar
  38. Lantuéjoul, C., Bacro, J.-N. and Bel, L. (2011) Storm Processes and Stochastic Geometry. Extremes 14(4), 413–428.Google Scholar
  39. Lindgren, F., Rue, H. and Lindström, J. (2011) An Explicit Link Between Gaussian Fields and Gaussian Markov Random Fields: The Stochastic Partial Differential Equation Approach. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 73(4), 423–498.Google Scholar
  40. Lindsay, B. G. (1988) Composite Likelihood Methods. Contemporary Mathematics 80, 221–239.Google Scholar
  41. Marcon, G., Padoan, S. A., Naveau, P. and Muliere, P. (2014) Multivariate Nonparametric Estimation of the Pickands Dependence Function Using Bernstein Polynomials. arXiv:1405.5228v2.
  42. Ng, C. T. and Joe, H. (2014) Model Comparison with Composite Likelihood Information Criteria. Bernoulli 20(4), 1738–1764.Google Scholar
  43. Nikoloulopoulos, A. K., Joe, H. and Li, H. (2009) Extreme Value Properties of Multivariate \(t\) Copulas. Extremes 12(2), 129–148.Google Scholar
  44. Northrop, P. J. and Jonathan, P. (2011) Threshold Modelling of Spatially-Dependent Non-Stationary Extremes with Application to Hurricane-Induced Wave Heights (with Discussion). Environmetrics 22, 799–809.Google Scholar
  45. Nychka, D., Wikle, C. K. and Royle, J. A. (2002) Multiresolution Models for Nonstationary Spatial Covariance Functions. Statistical Modelling 2(4), 315–331.Google Scholar
  46. Opitz, T. (2013) Extremal \(t\) Processes: Elliptical Domain of Attraction and a Spectral Representation. Journal of Multivariate Analysis 122(1), 409–413.Google Scholar
  47. Paciorek, C. J. and Schervish, M. (2006) Spatial Modelling Using a New Class of Nonstationary Covariance Functions. Environmetrics 17(5), 483–506.Google Scholar
  48. Padoan, S. A., Ribatet, M. and Sisson, S. A. (2010) Likelihood-Based Inference for Max-Stable Processes. Journal of the American Statistical Association 105(489), 263–277.Google Scholar
  49. Perrin, O. and Monestiez, P. (1999) Modelling of Non-Stationary Spatial Structure Using Parametric Radial Basis Deformations. In geoENV II – Geostatistics for Environmental Applications, eds J. Gómez-Hernández, A. Soares and R. Froidevaux, volume 10 of Quantitative Geology and Geostatistics, pp. 175–186. Springer.Google Scholar
  50. Reich, B. J., Eidsvik, J., Guindani, M., Nail, A. J. and Schmidt, A. M. (2011) A Class of Covariate-Dependent Spatiotemporal Covariance Functions For The Analysis of Daily Ozone Concentration. Annals of Applied Statistics 5(4), 2465–2487.Google Scholar
  51. Reich, B. J. and Shaby, B. A. (2012) A Hierarchical Max-Stable Spatial Model for Extreme Precipitation. Annals of Applied Statistics 6(4), 1430–1451.Google Scholar
  52. Ribatet, M. (2013) Spatial Extremes: Max-Stable Processes at Work. Journal de la Société Française de Statistique 154(2), 156–177.Google Scholar
  53. Ribatet, M., Cooley, D. S. and Davison, A. C. (2012) Bayesian Inference from Composite Likelihoods, with an Application to Spatial Extremes. Statistica Sinica 22(2), 813–845.Google Scholar
  54. Sampson, P. D. and Guttorp, P. (1992) Nonparametric Estimation of Nonstationary Spatial Covariance Structure. Journal of the American Statistical Association 87(417), 108–119.Google Scholar
  55. Schlather, M. (2002) Models for Stationary Max-Stable Random Fields. Extremes 5(1), 33–44.Google Scholar
  56. Schmidt, A. M. and O’Hagan, A. (2003) Bayesian Inference for Non-Stationary Spatial Covariance Structure via Spatial Deformations. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 65(3), 743–758.Google Scholar
  57. Smith, E. L. and Stephenson, A. G. (2009) An Extended Gaussian Max-Stable Process Model for Spatial Extremes. Journal of Statistical Planning and Inference 139(4), 1266–1275.Google Scholar
  58. Smith, R. L. (1990) Max-Stable Processes and Spatial Extremes. Unpublished.Google Scholar
  59. Stein, M. L. (1999) Interpolation of Spatial Data: Some Theory for Kriging. First edition. New York: Springer. ISBN 9780387986296.Google Scholar
  60. ——– (2005) Nonstationary Spatial Covariance Functions. Unpublished.Google Scholar
  61. Stephenson, A. and Tawn, J. A. (2005) Exploiting Occurrence Times in Likelihood Inference for Componentwise Maxima. Biometrika 92(1), 213–227.Google Scholar
  62. Thibaud, E., Aalto, J., Cooley, D. S., Davison, A. C. and Heikkinen, J. (2015) Bayesian Inference for the Brown–Resnick Process, With an Application to Extreme Low Temperatures. arXiv:1506.07836v1.
  63. Thibaud, E., Mutzner, R. and Davison, A. C. (2013) Threshold Modeling of Extreme Spatial Rainfall. Water Resources Research 49(8), 4633–4644.Google Scholar
  64. Thibaud, E. and Opitz, T. (2015) Efficient Inference and Simulation for Elliptical Pareto Processes. Biometrika 102(4), 855–870.Google Scholar
  65. Varin, C., Reid, N. and Firth, D. (2011) An Overview of Composite Likelihood Methods. Statistica Sinica 21, 5–42.Google Scholar
  66. Wadsworth, J. L. and Tawn, J. A. (2012) Dependence Modelling for Spatial Extremes. Biometrika 99(2), 253–272.Google Scholar
  67. ——– (2014) Efficient Inference for Spatial Extreme Value Processes Associated to Log-Gaussian Random Functions. Biometrika 101(1), 1–15.Google Scholar

Copyright information

© International Biometric Society 2016

Authors and Affiliations

  1. 1. CEMSE DivisionKing Abdullah University of Science and TechnologyThuwalSaudi Arabia

Personalised recommendations