Abstract
A flexible model called the max-mixture model has been introduced for modeling situations where the extremal dependence structure type may vary with distance. In this paper, we propose a novel estimation procedure for spatial max-mixture model parameters. Our procedure is based on the madogram, a dependence measure used in geostatistics to describe spatial structures. A nonlinear least squares minimization procedure is applied to obtain the estimators for extremal dependence functions. A simulation study shows that the proposed procedure works well for these models. In an analysis of monthly maxima of daily rainfall data collected over the East of Australia, we implement the proposed estimation procedure for diagnostic and confirmatory purposes.
Similar content being viewed by others
References
Abu-Awwad A-F, Maume-Deschamps V, Ribereau P (2019) Censored pairwise likelihood-based tests for mixture parameter of spatial max-mixture models. Accepted for publication to Revista de Investigacion Operacional. arXiv:1801.00981
Ahmed M, Maume-Deschamps V, Ribereau P, Vial C (2017) A semi-parametric estimation for max-mixture spatial processes. Submitted for publication
Antoniadis A, Berruyer J, René C (1992) Régression non linéaire et applications. Economica, Paris
Arbia G, Lafratta G (2002) Anisotropic spatial sampling designs for urban pollution. J R Stat Soc Ser C (Appl Stat) 51(2):223–234
Bacro JN, Gaëtan C, Toulemonde G (2016) A flexible dependence model for spatial extremes. J Stat Plan Inference 172:36–52
Beirlant J, Goegebeur Y, Segers J, Teugels J (2004) Statistics of extremes: theory and applications. Wiley, New York
Bel L, Bacro JN, Lantuéjoul C (2008) Assessing extremal dependence of environmental spatial fields. Environmetrics 19:163–182
Blanchet J, Davison AC (2011) Spatial modeling of extreme snow depth. Ann Appl Stat 5:1699–1725
Brown BM, Resnick SI (1977) Extreme values of independent stochastic processes. J Appl Probab 14:732–739
Buhl S, Davis RA, Klüppelberg C, Steinkohl C (2016) Semiparametric estimation for isotropic max-stable space-time processes. Submitted for publication
Capéraà P, Fougères AL, Genest C (1997) A nonparametric estimation procedure for bivariate extreme value copulas. Biometrika 84:567–577
Castruccio S, Huser R, Genton MG (2016) High-order composite likelihood inference for max-stable distributions and processes. J Comput Gr Stat 25:1212–1229
Coles S (2001) An introduction to statistical modeling of extreme values. Springer, London
Coles S, Heffernan J, Tawn J (1999) Dependence measures for extreme value analyses. Extremes 2:339–365
Cooley D, Naveau P, Poncet P (2006) Variograms for spatial max-stable random fields. In: Bertail P, Soulier P, Doukhan P (eds) Dependence in probability and statistics. Lecture notes in statistics, vol 187, chap 17. Springer, New York, pp 373–390
Davison AC, Gholamrezaee MM (2012) Geostatistics of extremes. Proc R Soc A 468:581–608
Davison AC, Huser R, Thibaud E (2013) Geostatistics of dependent and asymptotically independent extremes. Math Geosci 45:511–529
De Haan L (1984) A spectral representation for max-stable processes. Ann Probab 12:1194–1204
De Haan L, Pereira A (2006a) Extreme value theory: an introduction. Springer, New York
De Haan L, Pereira TT (2006b) Spatial extremes: models for the stationary case. Ann Stat 34:146–168
Ecker MD, Gelfand AE (2003) Spatial modeling and prediction under stationary non-geometric range anisotropy. Environ Ecol Stat 10(2):165–178
Genton MG, Ma Y, Sang H (2011) On the likelihood function of Gaussian max-stable processes. Biometrika 98:481–488
Guillou A, Naveau P, Schorgen A (2014) Madogram and asymptotic independence among maxima. REVSTAT Stat J 12(2):119–134
Huser R, Davison AC (2013) Composite likelihood estimation for the Brown–Resnick process. Biometrika 100:511–518
Huser R, Davison AC (2014) Space-time modelling of extreme events. J R Stat Soc Ser B (Stat Methodol) 76:439–461
Kabluchko Z, Schlather M, De Haan L (2009) Stationary max-stable fields associated to negative definite functions. Ann Probab 37:2042–2065
Ledford AW, Tawn JA (1996) Statistics for near independence in multivariate extreme values. Biometrika 83:169–187
Ledford AW, Tawn JA (1997) Modelling dependence within joint tail regions. J R Stat Soc Ser B (Stat Methodol) 59(2):475–499
Lindsay BG (1988) Composite likelihood methods. Contemp Math 80:221–239
Naveau P, Guillou A, Cooley D, Diebolt J (2009) Modelling pairwise dependence of maxima in space. Biometrika 96:1–17
Opitz T (2013) Extremal t processes: elliptical domain of attraction and a spectral representation. J Multivar Anal 122:409–413
Padoan SA, Ribatet M, Sisson SA (2010) Likelihood-based inference for max-stable processes. J Am Stat Assoc 105:263–277
Ribatet M (2015) SpatialExtremes: R Package, Version 2.0-2. http://spatialextremes.r-forge.rproject.org/. Accessed 2 July 2017
Schlather M (2002) Models for stationary max-stable random fields. Extremes 5:33–44
Schlather M, Tawn JA (2003) A dependence measure for multivariate and spatial extreme values: properties and inference. Biometrika 90:139–156
Sibuya M (1960) Bivariate extreme statistics. Ann Inst Stat Math 11:195–210
Smith RL (1990) Max-stable processes and spatial extremes. Unpublished manuscript 205
Thibaud E, Mutzner R, Davison AC (2013) Threshold modeling of extreme spatial rainfall. Water Resour Res 49:4633–4644
Varin C (2008) On composite marginal likelihoods. AStA Adv Stat Anal 92:1–28
Varin C, Vidoni P (2005) A note on composite likelihood inference and model selection. Biometrika 92:519–528
Varin C, Reid N, Firth D (2011) An overview of composite likelihood methods. Stat Sin 21:5–42
Wadsworth JL, Tawn JA (2012) Dependence modelling for spatial extremes. Biometrika 99:253–272
Acknowledgements
This work has been supported by the LABEX MILYON (ANR-10-LABX-0070) of Université de Lyon, within the program “Investissements d’Avenir” (ANR-11-IDEX-0007) operated by the French National Research Agency (ANR). It was also partly supported by the CERISE LEFE-INSU project. We would like to thank Jean-Noël Bacro, Carlo Gaetan, and Gwladys Toulemonde for providing their C codes which used as a preliminary version for parametric estimation using composite likelihood. We are also very grateful to an editor and to two anonymous reviewers for many helpful comments that have led to considerable improvements in the manuscript.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Abu-Awwad, A., Maume-Deschamps, V. & Ribereau, P. Fitting spatial max-mixture processes with unknown extremal dependence class: an exploratory analysis tool. TEST 29, 479–522 (2020). https://doi.org/10.1007/s11749-019-00663-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11749-019-00663-5
Keywords
- Asymptotic dependence
- Asymptotic independence
- Composite likelihood
- Madogram
- Max-stable model
- Max-mixture model
- Nonlinear least squares